System and Method for Monitoring Train Properties and Maintenance Quality

Field

The invention lies in the field of railways monitoring and particularly in the field of analyzing train properties of railway components. More particularly, the present invention relates to a system for analyzing properties of trains and their maintenance, a method performed in such a system and corresponding use of the system.

Introduction

Railroad, railway or rail transport has been developed for transferring goods and passengers on wheeled vehicles on rails, also known as tracks. In contrast to road transport, where vehicles run on a prepared flat surface, rail vehicles (rolling stock) are directionally guided by the tracks on which they run. Tracks commonly consist of steel rails, installed on ties or sleepers and ballast, on which the rolling stock, usually provided with metal wheels, moves. Other variations are also possible, such as slab track, where the rails are fastened to a concrete foundation resting on a subsurface.

Rolling stock in a rail transport system generally encounters lower frictional resistance than road vehicles, so passenger and freight cars (carriages and wagons) can be coupled into longer trains. Power is provided by locomotives, which either draw electric power from a railway electrification system or produce their own power, usually by diesel engines. Most tracks are accompanied by a signaling system. Railways are a safe land transport system when compared to other forms of transport. Additionally, railways are capable of high levels of passenger and cargo utilization and energy efficiency but are often less flexible and more capital-intensive than road transport, when lower traffic levels are considered.

The inspection of railway equipment is essential for the safe movement of trains. Many types of defect detectors are in use today. These devices utilize technologies that vary from a simplistic paddle and switch to infrared and laser scanning, and even ultrasonic audio analysis. Their use has avoided many rail accidents over the past decades.

Railway operations require careful monitoring and control of the conditions of the railway component to ensure passenger safety and reliable service. Many sensors are used to monitor and obtain data from different infrastructural components of the railway network, which may be used to ensure the integrity of the service and identify possible sources of malfunction. Such sensors allow for data collection and analysis and ensure safer operations of railways. Various sensors can be placed directly on trains, on tracks or nearby, at train stations and/or on platforms, and generally in the overall vicinity of the railway. In particular, such sensors may be used to obtain acceleration patterns, which may be induced by a passing train and measured, for instance, by an accelerometer affixed to the rail or sleeper. Acceleration patterns may contain information about properties of the train that induced said vibrations, including speed, weight, wheel health, and axle patterns. However, data collection is technically complex as, for example, for weight data it is required measurement of a weight station or other source of ground truth to be correlated with acceleration signals, which limits severely its applicability.

For example, Bruni, S., et al. discloses effects of train impacts on urban turnouts: modelling and validation through measurements as train-track interaction at turnouts is a main issue in the design and maintenance of railway systems. Due to the particular geometry of wheel-rail contact and to the sudden variation of track flexibility, severe impact loads may occur during train passage over the turnout. In this paper, two different modelling approaches to reproduce train-turnout interaction are proposed and compared. A first technique, developed by Politecnico di Milano, is based on a detailed multi-body model of the trainset and of wheel-rail contact, whereas for the turnout structure a simplified finite element model is used. The second modelling technique, developed by the National Technical University of Athens, relies on a detailed finite element model of the turnout, while a simplified model is used to compute impact loading due to wheel passage. In this paper the two methods are validated trough comparison with line measurements performed on turnout systems from three different urban railway networks.

Evans, J and Berg, M also discloses challenges in simulation of rail vehicle dynamics as rail vehicle dynamic simulation has progressed a long way from its origins as a research tool. Modern multibody software packages are used as an essential part of the design process for new vehicles and for investigating service problems with existing vehicles. Increasingly, simulation is being used as part of the vehicle acceptance process in place of on-track testing. This state-of-the-art paper for the 21st IAVSD Symposium in Stockholm in August 2009 surveys the current applications for rail vehicle dynamic modelling. The process of reducing a complex mechanical system to a mathematical representation is invariably subject to compromise and open to individual interpretation. The level of detail and choice of idealization of suspension components will depend on the application, and in the real world it also depends on the availability of information about the system. This paper discusses appropriate modelling choices for different applications, and comments on best practice for the idealization of suspension components, wheel/rail contact conditions and modelling inputs such as track geometry. The validation of simulation results is increasingly important, and this paper discusses recent trends in this area. Finally, the paper takes a brief look forward to future simulation issues. Although all approaches output different values, they all face similar problems related to productivity and scalability. Additionally, when data are collected from multiple sources a cross validation is necessary to yield accurate predictions. Another disadvantage of the existing methods is the disability to handle nested data.

Summary

In light of the above, it is an object of the present invention to overcome or at least alleviate the shortcomings of the prior art. More particularly, it is an object of the present invention to provide a system and a method for analyzing physical properties of railway components comprised by the railway network.

It is also a preferred object of the present invention monitoring health status of a railway network, in particular, components of a railway component comprised by the railway network.

These objects are met by the present invention wherein a system for monitoring properties of at least one train is disclosed. The system comprises at least one sensor component configured to sample at least one sensor data relevant to the at least one train. The system further comprises at least one processing component configured to process the at least one sensor data. The system comprises at least one storing component configured to store the at least one sensor data, and at least one analyzing component.

In some embodiments the analyzing component can be configured to receive the at least one sensor data from the at least one sensor component. The system can further be configured to receive at least one signal comprising a frequency range between 0 and 10,000 Hz, preferably between 0 and 8,000 Hz, more preferably between 0 and 2,000 Hz. The receiving frequency in such embodiments can be the vibrational frequency caused by accelerating or deaccelerating trains.

In some embodiments the processing component can be configured to receive the at least one signal comprising a frequency range between 0 and 10,000 Hz, preferably between 0 and 8,000 Hz, more preferably between 0 and 2,000 Hz. In such embodiments the system can be configured to estimate at least one estimation value of the at least one train, preferably based on the frequency. In some embodiments the at least one sensor component can be configured to at least partially determine the at least one estimation value of the at least one train. This can be particularly advantageous if the sensor component is installed at the vibration site and/or within a distance of 0.1 to 2 m from the railway tracks. The system can further be configured to estimate the at least one estimation value of at least one property of the at least one train. It should be noted The property can comprise a geometric property, locomotive property, kinematic property, physical property, etc. In some embodiments the analyzing component can be configured to at least partially determine the at least one estimation value of the at least one train. In such embodiments the analyzing component may be equipped with at least one machine learning technique. In some embodiments the analyzing component can be facilitated with digital twin technology.

In some embodiments the analyzing component can be configured to at least partially associate the at least one estimation value to an interaction between at least 2 railway components of the railway network. In such embodiments the analyzing component can further be configured to monitor at least one railway health status of at least one component of the railway network. In such embodiments the analyzing component can be configured to forecast at least one railway health status of at least one component of the railway network.

In some embodiments the analyzing component is configured to generate at least a first level model. In such embodiments the first level model may comprise a hierarchical linear model and/or linear mixed-effect model and/or nested data model and/or random coefficient model and/or random effects model and/or random parameter model and/or split-plot designs. In the first level model. The first level model may comprise a level 1 regression equation further establishing a dependency of the at least two properties/features of railway infrastructure.

In some embodiments the first level model can be based the at least one estimation value. In some further embodiments the at least first level model can be generated based on at least one estimated value. The estimated value comprises at least one physical, chemical and/or environmental parameter(s) in railway infrastructure and/or the train. In some embodiments the system can be further configured to calculate a first interaction score between at least two estimated values.

In some embodiments the system and/or the analyzing component can be configured to generate at least one second level model. In such embodiments the second level model may comprise a hierarchical linear model and/or linear mixed-effect model and/or nested data model and/or random coefficient model and/or random effects model and/or random parameter model and/or split-plot designs. The second level model may comprise a level 2 regression equation. In some embodiments the second level model can be configured to be generated based on the at least first interaction score. The interaction score can comprise a coefficient associated with the at least one parameter, where the parameter may be a representation of the property of railway infrastructure and/or train. The second level model can be configured to be generated based on the at least one estimated value. The system can be further configured to generate a second interaction score.

In some embodiments the second interaction score is configured to be generated based on the second level model. The second interaction score can be configured to be based on the at least one estimated value. The second interaction score can be configured to be based on at least one variable. The variable can comprise at least one physical, chemical and/or environmental variable in railway infrastructure and/or train.

In some embodiments the system is further configured to calibrate the at least first level model based on ground truth. In such embodiments ground truth can comprise information/equation/parameters that is known to be real or true, provided by direct observation and measurement (i.e., empirical evidence). The system is further configured to calibrate the at least second level model based on ground truth. Further, the analyzing component can be configured to train the at least one machine learning model using the at least first level model.

In some embodiments the analyzing component can further be configured to train the at least one machine learning model using the at least second level model. In such embodiments the sensor data can comprise nested sensor data. The nested sensor data can comprise the data set which can be assigned to hierarchically superior units.

In some embodiments the system cans further be configured to generate at least one third level model. The third model can comprise a hierarchical linear model and/or linear mixed- effect model and/or nested data model and/or random coefficient model and/or random effects model and/or random parameter model and/or split-plot designs. In such embodiments the third level model can be configured to be based on the at least second interaction score. The third level model can be configured to be based on the at least first interaction score. The analyzing component can further be configured to determine the at least one estimated value based on the third level model. The analyzing component can further be configured to determine the at least one variable based on the third level model. Furthermore, the analyzing component can be configured to be stored on a server.

In some embodiments the analyzing component can be configured to label the sensor data based on the at least first level model. The analyzing component can be configured to label the sensor data based on the at least second level model. The analyzing component can also be configured to label the sensor data based on the at least third level model. Furthermore, the labelled data can be used to train the machine learning algorithm thus health monitoring of railway infrastructure/trains.

In some embodiments the system can be configured to execute a multilevel modeling. The system and/or the analyzing component can be configured to generate a first level model. The first level model can be based on the at least one estimation value. The estimation value can be a representation of a feature associated with railway infrastructure. The system can be configured to estimate the at least one estimation value of the at least one train, preferably based on the frequency. The system can further be configured to estimate the at least one estimation value of at least one property of the at least one train. The system can further be configured to estimate the at least one estimation value of the railway infrastructure. It should be noted that the property can comprise a geometric property, locomotive property, kinematic property, physical property, etc. The analyzing component can be configured to at least partially determine the at least one estimation value of the at least one train. In general estimation value can further comprises processed sensor data. The multilevel modeling as shown in this exemplary embodiment can be used when the sensor data is nested data.

In some further embodiments the first level model can also be based on at least one estimated value. The estimated value can be an inferred estimation value. The estimated value can comprise, for example, a weight estimation of at least one component in the railway infrastructure. In such embodiments the railway infrastructure can comprise the train, such as weight estimation may comprise weight estimation of the train. The estimated value can further comprise an axle patterns estimation or the like. The estimated value and/or the estimation value can comprise at least one physical, chemical and/or environmental parameter in railway infrastructure. The first level model can be configured to provide a weighted average of pooled model estimates. The multilevel modeling can further comprise Bayesian hierarchical modeling. This modeling can particularly be advantageous when the at least two estimated values are independent of each other. The estimated values can be nested within one another at different levels.

In some embodiments the system can further be configured to generate at least a first interaction score. The first interaction score can comprise a coefficient, such as a regression coefficient associating the at least two estimated values. For example, the sensor data can comprise acceleration data of a train, in some embodiments a glitch in acceleration value can imply a void in the rocks, in such embodiment the model may generate the first interaction score associating the glitch to the void. The first interaction score can comprise a random coefficient, random parameter model, nested data models and/or split plot designs. The system can be configured to generate a second level model. The second level model can be configured to be based on the at least first interaction score.

In some embodiments the second level model can further be configured to generate a second interaction score. The second interaction score can be based on the variable. The variable can comprise at least one regression coefficient. The variable can further comprise the error component. In some embodiments the variable can further comprise a hyperprior. For example, a second level model can comprise the second interaction score associating the rail-wheel contact force with the weight of the train. And based on this may calculate the dip angle on the tracks.

In some embodiments the system can further generate a third level model, wherein the third level model can be based on the second interaction score. The third level model may indicate a position on the tracks where the acceleration is increased. It is to be noted that the term railway infrastructure can also comprise the vehicle in the railway structure, such as the train. In general, where ever the term train properties are mentioned this can comprise properties of the vehicle in a railway infrastructure, such as speed of the train, axel pattern, category of the train (passenger train, cargo, etc.), weight of the train, properties related to wheels of the train (flat and/or deformed), etc. Additionally, and/or optionally, the term railway infrastructure can comprise geometric and/or physical properties of the infrastructure, such as stiffness and dampening of subsoil, ballast or rail pads, presence of void under sleeper, eigenfrequencies of sleepers, presence of surface defects on rail, dip angle of frig, geometry etc.

In some further embodiments the analyzing component can be configured to generate at least one railway health status hypothesis comprising at least one cause for the at least one railway health status of the at least one component of the railway network. The railway health status can comprise health assessment of components of railway infrastructure. The railway health status hypothesis can comprise fault detection, location, severity, classification and prognosis based on the sensor data. The at least one sensor data relevant to the railway network can comprise at least one railway infrastructural feature.

In some embodiments the analyzing component can comprise a self-learning module. The self-learning module can comprise at least one of supervised and unsupervised and reinforcement learning training modules. The self-learning module can also be configured to analyze the at least one sensor data. Further, the self-learning module is configured to analyze at least one of the at least one estimation value, and the at least one property of the at least one train. In some embodiments the self-learning module can be configured to determine changes over time of at least one of the at least one sensor data, the at least one estimation value, and the at least one property of the at least one train. The self-learning module can further be configured to correlate changes over time with at least one railway health status hypothesis. The railway health hypothesis can comprise the estimation value to be in a threshold range. For example, if frequency in the range of 20-100 Hz is recorded when an ICE passes from a location it could mean that the ballast at the location is deformed and needs to be replaced. In some further embodiments the analyzing component can be configured to generate the railway health hypothesis taking into account noise. In some embodiments the noise can also be taken into account before the sensor data is fed into the analyzing component. The noise can comprise wheel-rail noise, rolling noise, aerodynamic noise, impact noise, squeal, structure-borne noise, ground borne noise, viaducts, etc.

In some embodiments the self-learning module can further be configured to execute at least one simulation model. In some embodiments the system can further comprise at least one computing component. The computing component can be a remote computing component. The computing component can be a local computing component. The at least one computing component can be integrated in the at least one sensor component. The at least one computing component can further be configured to pre-process the at least one sensor data. The at least one computing component can further comprises the self-learning module. In some embodiments the analyzing component can be configured to execute at least one analytical approach. The analytic approach can comprise at least one of: signal filter processing, pattern recognition, probabilistic modeling, Bayesian methods, machine learning, supervised learning, unsupervised learning, reinforcement learning, statistical analytics, statistical models, principle component analysis, independent component analysis (ICA), dynamic time warping, maximum likelihood estimates, modeling, estimating, neural network, convolutional network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithms, particle filters, variations of Kalman filters, Markov models, and/or hidden Markov models.

In some embodiments the system can further comprise at least one server. The at least one server can comprise a remote server. The at least one server can comprise a local server. The at least one server can further be configured to receive at least one sensor data relevant to the railway network. In some embodiments the at least one server can be configured to monitor at least one of the at least one sensor data, the at least one estimation value, and the at least one property of the at least one train. In some embodiments the at least one server can be configured to provide at least one signal, wherein the signal can comprise the at least one sensor data. In such embodiments the at least one signal can comprise the at least one estimation value. Further, the at least one signal can comprise the at least one property of the at least one train. The at least one server can further be configured to provide at least one signal which can comprise at least one railway infrastructural data. The at least one signal can be processed based on at least one analytical approach. The at least one sensor can be configured to operate in a plurality of operation modes. Further, each operation mode can be configured to monitor at least one sensor data relevant to railway network. It is to be noted that here the operation mode can comprise at least one of sensor, server, analyzing component.

In some further embodiments the at least one server can comprise an interface module. In such embodiments the server can be configured to bidirectionally communicate with at least one authorized user. The interface module can be configured to prompt at least one action by the at least one authorized user. Furthermore, the at least one server can be configured to forecast health status of the at least one railway network based on at least one of the at least one eigenfrequency, the at least one property of the at least one train and the at least one sensor data.

In some embodiments the at least one storing component can be configured to store data generated by the at least one server. The at least one railway component can comprise at least one unmovable component, such as railway tracks. The at least one railway component can further comprise at least one movable component such as railway switches, frogs, rail barriers. In some embodiments the processing component can be configured to generate structured datasets using the at least one sensor data. Further, the processing component can be configured to execute at least one analytical approach.

In some embodiments the analyzing component can be configured to execute the at least one analytical approach. In some embodiments the at least one server comprises at least one of the at least one processing component, and the at least one analyzing component. The system can further be configured to execute at least one machine learning algorithm, such as pattern recognition. It may be noted here that the machine learning used can be any classifier such as random forest classifier. In some embodiments heuristic search algorithms can be used. The at least one processing component can be configured to (automatically) perform signal processing, for example using sophisticated FFT.

In some embodiments the at least one sensor can comprise an acceleration sensor. In some embodiments sensor can be velocity sensor. In some embodiments the sensor can comprise a displacement sensor. In some further embodiments the sensor can comprise a pressure sensor. It should be noted that in some further embodiments the at least one sensor can comprise any combination(s) of the sensors mentioned above. In some embodiments the sensor can be configured to measure data in a vertical direction. Here the vertical direction can comprise the direction in which the vehicle is moving relative to the sensor. In some embodiments the sensor can be configured to measure data in a longitudinal direction. In some embodiments the at least one sensor can be configured to measure data in a lateral direction.

In some embodiments the estimation value comprises a risk estimation value, such as risks associated with a hazard. In such embodiments hazard can be at least one of the following the system boundary and how it interacts with its environment, the operational function of the system, including normal/ degraded and emergency modes, the life cycle of the system, including any necessary maintenance, the operating functions (what the system will interact with), human influence, the operational environment, reasonably foreseeable failure modes.

In some embodiments the system can be configured to build the at least one simulation model. It is to be understood that the term build here implies construction of a simulation model in a virtual environment, preferably by an algorithm. The simulation model can further be configured to comprise a vehicle-track interaction in a multi body simulation environment. This can facilitate calculation of sleeper-ballast contact pressures and/or sleeper acceleration. In some embodiments this can be accomplished by assuming a starting value for pad stiffness and ballast stiffness. It should be noted that the stiffness of the component in the railway infrastructure can be measure by using existing physical means.

In some further embodiments the system can execute the simulation model to extract at least sleeper acceleration. In some further embodiments the analyzing component can extract the sleeper contact pressure. In some embodiments the system can further be configured to build the simulation model which can be configured based on a track settlement model. In such embodiments the simulation model can be configured to be built using time series analysis assuming that the ballast is uncompressed at time=0 sec.

In some embodiments the system can further be configured to calculate at least permanent deformation in at least ballast geometry based on the track settlement model. In some embodiments the system can be configured to generate a rail pad model. In some further embodiments the analyzing component can be configured to generate the rail pad model. In such embodiments the rail pad model can be used to measure static stiffness of the rail pads as a function of time. In some further embodiments the analyzing component can be configured to calculate dynamic stiffness and/or damping of the rail pad using the rail pad model. In some further embodiments the simulation model can comprise the rail pad model.

In some embodiments the analyzing component can further be configured to measure an influence of the train speed on fatigue criterion of the railway tracks. This can be done automatically using the at least one simulation model. In some embodiments the analyzing component can further configured to measure an influence of connection rail and sleeper, which can also be accomplished automatically. Furthermore, the analyzing component can be configured to measure an influence of the weight of the train on the tracks and/or to measure an influence of the wear of the wheels on the speed of the train.

In a second embodiment a method for monitoring properties of at least one train is disclosed. The method comprising collecting at least one sensor data at least one time via a least one sensor arranged on at least one railway component. Further the method comprising determining at least one property of the at least one train based on the at least one sensor data, and generating at least one determined property finding.

The method can comprise inferring an estimation value of the at least one property of the at least one train, and/or generating at least one estimated value and/or calibrating a physical model of the at least one property and/or generating at least one calibrated physical model. In some embodiments the method also comprise calibrating of the physical model is based on the at least one determined property finding. In such embodiments calibrating of the physical model can be based on the at least one estimated value. The at least one estimated value can comprise a weight estimation. In some further embodiments the at least one estimated value can comprise an axle patterns estimation.

In some embodiments the at least one estimated value can comprise a train speed estimation. In some further embodiments the estimated value can comprise a wheel health estimation. Furthermore, the step of estimating an estimation value of the at least one property of the at least one train can be based on at least one undedicated measurement, wherein the undedicated measurement can comprise measurements which can be made without the need of dedicated measuring devices, like a weight station, or external data to build a labelled sample set.

In some further embodiments the step of calibrating of the physical model can comprise calibrating the physical model to at least one acceleration data. Further, calibrating of the physical model can comprise calibrating the physical model to at least one estimation of train weight. The calibrating of the physical model can further comprise calibrating the physical model to at least one speed data. In some embodiments calibrating of the physical model can comprise calibrating the physical model to at least one wheel health data. The calibrating of the physical model can comprise calibrating the physical model to at least one axle pattern data. The method can comprise monitoring at least one physical parameter of at least one asset based on the at least one calibrated physical model. The method can further comprise generating at least one corrective measure, and applying the at least one corrective measure to the asset. The method can comprise evaluating an effectiveness of the at least corrective measure.

In some embodiments the method can comprise performing a continuous monitoring, wherein the continuous monitoring comprises providing a series of measurements of the at least one physical parameters, wherein the method comprises an initial measurement and at least one subsequent measurement, wherein the method comprises comparing initial measurement and the at least one subsequent measurement to generate at least one evolution status, wherein the at least one evolution status is based on health status of the asset before and after a corrective measure. The method can comprise determining an outcome status of the at least one corrective measure. Additionally, or optionally, the method can comprise quantifying a degree of effectiveness of the at least one corrective measure. In such embodiments the degree of effectiveness can be quantifying based on the outcome status of the at least one corrective measure. For example, tamping is often done for two main reasons: adjusting geometry or for vertical displacement control, e.g. removing a void under a sleeper. These aspects would be part of the monitored digital twin, and would therefore allow to infer the state of them after the maintenance action is undertaken, but also it is almost always the case that after an initial improvement the asset settles, and the condition deteriorates, if not to the original bad place, at least somewhat. Monitoring that we would be able to tell not only if the maintenance activity had an immediate effect, but also how effective it was over time. Infrequent measurements do not allow for this. Currently what an asset operator often does is to measure at spread out times either the direct physical property, e.g., geometry, or proxies for properties of interest, e.g., displacement. However, because that measurement was not performed frequently enough before and after the maintenance action, it is not always possible to infer the causal effect of the maintenance action, nor ascertain the effectiveness in correcting the underlying problem.

The system can further be configured to perform the method according to any of the preceding method embodiments.

The present technology is also defined by the following numbered embodiments. Below, system embodiments will be discussed. These embodiments are abbreviated by the letter "S" followed by a number. Whenever reference is herein made to "system embodiments", these embodiments are meant.

51. A system for monitoring properties of at least one train, the system comprising at least one sensor component configured to sample at least one sensor data, at least one processing component configured to process the at least one sensor data, at least one storing component configured to store the at least one sensor data, and at least one analyzing component.

52. The system according to the preceding embodiment, wherein the at least one analyzing component is configured to receive the at least one sensor data from the at least one sensor component.

53. The system according to any of the preceding embodiments, wherein the system is configured to receive at least one signal comprising a frequency range between 0 and 10,000 Hz, preferably between 0 and 8,000 Hz, more preferably between 0 and 2,000 Hz.

54. The system according to any of the preceding embodiments, wherein the processing component is configured to receive at least one signal comprising a frequency range between 0 and 10,000 Hz, preferably between 0 and 8,000 Hz, more preferably between 0 and 2,000 Hz.

55. The system according to any of the preceding embodiments, wherein the system is further configured to estimate at least one estimation value of the at least one train.

56. The system according to the preceding embodiment, wherein the at least one sensor component is configured to at least partially determine the at least one estimation value of the at least one train.

57. The system according to any of the preceding embodiments, wherein the system is configured to estimate the at least one estimation value of at least one property of the at least one train.

58. The system according to any of the 2 preceding embodiments, wherein the at least one analyzing component is configured to at least partially determine the at least one estimation value of the at least one train. S9. The system according to any of the 3 preceding embodiments, wherein the at least one analyzing component is configured to at least partially associate the at least one estimation value to an interaction between at least 2 railway components of the railway network.

510. The system according to any of the preceding embodiments wherein the analyzing component is configured to generate at least a first level model.

511. The system according to the preceding embodiment wherein the first level model is based on the at least one estimation value.

512. The system according to any of the preceding embodiments wherein the at least first level model is generated based on at least one estimated value.

513. The system according to the preceding embodiment wherein the estimated value comprises at least one physical, chemical and/or environmental parameter in railway infrastructure.

514. The system according to any of the preceding embodiments wherein the system is further configured to calculate a first interaction score between at least two estimated values.

515. The system according to any of the preceding embodiments wherein the system is configured to generate at least one second level model.

516. The system according to any of the preceding embodiments wherein the at least one portion of the second level model is configured to be generated based on the at least first interaction score.

517. The system according to any of the preceding embodiments wherein the at least one portion of second level model is configured to be generated based on the at least one estimated value.

518. The system according to any of the preceding embodiments wherein the system is further configured to generate a second interaction score.

S19. The system according to any of the preceding embodiments wherein the second interaction score is configured to be generated based on the second level model. 520. The system according to any of the preceding embodiments wherein the second interaction score is configured to be based on the at least one estimated value.

521. The system according to any of the preceding embodiments wherein the second interaction score is configured to be based on at least one variable.

522. The system according to the preceding embodiment wherein the variable comprises at least one physical, chemical and/or environmental variable in railway infrastructure.

523. The system according to any of the preceding embodiments wherein the system is further configured to calibrate the at least first level model based on ground truth.

524. The system according to any of the preceding embodiments wherein the system is further configured to calibrate the at least second level model based on ground truth.

525. The system according to any of the preceding embodiments wherein the analyzing component is further configured to train the at least one machine learning model using the at least first level model.

526. The system according to any of the preceding embodiments wherein the analyzing component is further configured to train the at least one machine learning model using the at least second level model.

527. The system according to any of the preceding embodiments wherein the sensor data comprises nested sensor data.

528. The system according to any of the preceding embodiments wherein the system is further configured to generate at least one third level model.

529. The system according to any of the preceding embodiments wherein the third level model is configured to be based on the at least second interaction score.

530. The system according to any of the preceding embodiments wherein the third level model is configured to be based on the at least first interaction score.

S31. The system according to any of the preceding embodiments wherein the analyzing component is further configured to determine the at least one estimated value based on the third level model. S32. The system according to any of the preceding embodiments wherein the analyzing component is further configured to determine the at least one variable based on the third level model.

533. The system according to any of the preceding embodiments wherein the analyzing component is further configured to be stored on a server.

534. The system according to the preceding embodiment wherein the analyzing component is configured to label the sensor data based on the at least first level model.

535. The system according to any of the preceding embodiments wherein the analyzing component is configured to label the sensor data based on the at least second level model.

536. The system according to any of the preceding embodiments wherein the analyzing component is configured to label the sensor data based on the at least third level model.

537. The system according to any of the 3 preceding embodiments, wherein the at least one analyzing component is configured to monitor at least one railway health status of at least one component of the railway network.

538. The system according to any of the 4 preceding embodiments, wherein the at least one analyzing component is configured to forecast at least one railway health status of at least one component of the railway network.

539. The system according to any of the preceding embodiments, wherein the at least one analyzing component is configured to generate at least one railway health status hypothesis comprising at least one cause for the at least one railway health status of the at least one component of the railway network.

540. The system according any of the preceding embodiments, wherein the at least one sensor data relevant to the railway network comprises at least one railway infrastructural feature.

541. The system according to any of the preceding embodiments, wherein the system, preferably at least one analyzing component comprises a self-learning module.

542. The system according to the preceding embodiment, wherein the self-learning module is configured to analyze the at least one sensor data. 543. The system according to embodiment S12, wherein the self-learning module is configured to analyze at least one of the at least one estimation value, and the at least one property of the at least one train/railway infrastructure.

544. The system according to any of the 3 preceding embodiments, wherein the selflearning module is configured to determine changes over time of at least one of the at least one sensor data, the at least one estimation value, and the at least one property of the at least one train/railway infrastructure.

545. The system according to any of the 4 preceding embodiments, wherein the selflearning module is configured to correlate changes over time with at least one railway health status hypothesis.

546. The system according to any of the preceding embodiments wherein the railway health hypothesis comprises the estimation value to be in a threshold range.

547. The system according to any of the preceding embodiments wherein the analyzing component is configured to generate the railway health hypothesis after calculating noise.

548. The system according to any of the 5 preceding embodiments, wherein self-learning module is further configured to execute at least one simulation model.

549. The system according to any of the preceding embodiments, wherein the system comprises at least one computing component.

550. The system according to the preceding embodiment, wherein the at least one computing component is a remote computing component.

551. The system according to embodiment S19, wherein the at least one computing is a local computing component.

552. The system according to embodiment S49 or S50, wherein the at least one computing component is integrated in the at least one sensor component.

S53. The system according to any of the 4 preceding embodiments wherein the at least one computing component is configured to pre-process the at least one sensor data. 554. The system according to any of the 5 preceding embodiments wherein the at least one computing component comprises the self-learning module.

555. The system according to any of the preceding embodiments, wherein the at least one analyzing component is configured to execute at least one analytical approach.

556. The system according to any of the preceding embodiments, wherein the system further comprises the at least one server.

557. The system according to the preceding embodiment, wherein the at least one server comprises a remote server.

558. The system according to any of the 2 preceding embodiments, wherein the at least one server comprises a local server.

559. The system according to the 3 preceding embodiments, wherein the at least one server is configured to receive at least one sensor data relevant to the railway network.

560. The system according to the 4 preceding embodiments, wherein the at least one server is configured to monitor at least one of the at least one sensor data, the at least one estimation value, and the at least one property of the at least one train.

561. The system according to any of the preceding embodiments wherein the at least one server is configured to provide at least one signal.

562. The system according to the preceding embodiment, wherein the at least one signal comprises the at least one sensor data.

563. The system according to embodiment S61, wherein the at least one signal comprises the at least one estimation value.

564. The system according to embodiment S61, wherein the at least one signal comprises the at least one property of the at least one train.

S65. The system according to any of the preceding embodiments wherein the at least one server is configured to provide at least one signal comprising at least one railway infrastructural data. 566. The system according to any of the preceding embodiments, wherein the at least one signal is processed based on at least one analytical approach.

567. The system according to any of the preceding embodiments, wherein the at least one sensor is configured to operate in a plurality of operation modes, and wherein each operation mode is configured to monitor at least one sensor data relevant to railway network.

568. The system according to any of the preceding embodiments, wherein the at least one server comprises an interface module configured to bidirectionally communicate with at least one authorized user.

569. The system according to the preceding embodiment, wherein the interface module is configured to prompt at least one action by the at least one authorized user.

570. The system according to any of the preceding embodiments, wherein the at least one server is configured to forecast health status of the at least one railway network based on at least one of the at least one eigenfrequency, the at least one property of the at least one train and the at least one sensor data.

571. The system according to any of the preceding embodiments, wherein the at least one storing component is configured to store data generated by the at least one server.

572. The system according to the preceding embodiment, wherein the at least one railway component comprises at least one unmovable component, such as railway tracks.

573. The system according to any of the 2 preceding embodiments, wherein the at least one railway component comprises at least one movable component such as railway switches, frogs, rail barriers.

574. The system according to any of the preceding embodiments, wherein the processing component is configured to generate structured datasets using the at least one sensor data.

S75. The system according to any of the preceding embodiments, wherein the processing component is configured to execute at least one analytical approach. S76. The system according to any of the preceding system embodiments, wherein the analyzing component is configured to execute the at least one analytical approach.

S77. The system according to any of the 2 preceding embodiments and with features of embodiment S26, wherein at least one server comprises at least one of the at least one processing component, and the at least one analyzing component.

578. The system according to any of the 3 preceding embodiments, wherein the system is configured to execute at least one machine learning algorithm.

579. The system according to the preceding embodiment, wherein the at least one machine learning algorithm comprises pattern recognition.

580. The system according to any of the preceding embodiments, wherein the at least one processing component is configured to (automatically) perform signal processing.

581. The system according to any of the preceding embodiments wherein the at least one sensor comprises an acceleration sensor.

582. The system according to any of the preceding embodiments wherein the at least one sensor comprises a velocity sensor.

583. The system according to any of the preceding embodiments wherein the at least one sensor comprises a displacement sensor.

584. The system according to any of the preceding embodiments wherein the at least one sensor comprises a pressure sensor.

585. The system according to any of the preceding embodiments, wherein the at least one sensor is configured to measure data in a vertical direction.

586. The system according to any of the preceding embodiments, wherein the at least one sensor is configured to measure data in a longitudinal direction.

587. The system according to any of the preceding embodiments, wherein the at least one sensor is configured to measure data in a lateral direction. S88. The system according to any of the preceding embodiments wherein the estimation value comprises a risk estimation value, such as risks associated with a hazard.

589. The system according to any of the preceding embodiments wherein the system is configured to build the at least one simulation model.

590. The system according to any of the preceding embodiments wherein the simulation model is further configured to comprise a vehicle-track interaction in a multi-body simulation environment.

591. The system according to any of the preceding embodiments wherein the system is further configured to execute the simulation model to extract at least sleeper-ballast contact pressure.

592. The system according to any of the preceding embodiments wherein the analyzing component is further configured to execute the simulation model to extract at least sleeperballast contact pressure.

593. The system according to any of the preceding embodiments wherein the system is further configured to execute the simulation model to extract at least sleeper acceleration.

594. The system according to any of the preceding embodiments wherein the analyzing component is further configured to execute the simulation model to extract at least sleeperballast contact pressure.

595. The system according to any of the preceding embodiments wherein the system is further configured to build the simulation model further configured based on a track settlement model.

596. The system according to any of the preceding embodiments wherein the system is further configured to build the simulation model based on the track settlement model, wherein the simulation model is configured to be built using time series analysis assuming that the ballast is uncompressed at time=0 sec.

597. The system according to any of the preceding embodiments wherein the system is configured to calculate at least permanent deformation in at least ballast geometry based on the track settlement model. S98. The system according to any of the preceding embodiments wherein the system is configured to generate a rail pad model.

S99. The system according to any of the preceding embodiments wherein the analyzing component is configured to generate the rail pad model.

5100. The system according to any of the preceding embodiments wherein the system is configured to generate the rail pad model, wherein the rail pad model is used to measure a static stiffness of the rail pads as a function of time.

5101. The system according to any of the preceding embodiments wherein the system is configured to generate the rail pad model, wherein the analyzing component is configured to calculate dynamic stiffness and/or damping of the rail pad using the rail pad model.

5102. The system according to any of the preceding embodiments wherein the analyzing component is further configured to measure an influence of the train speed on fatigue criterion of the railway tracks.

5103. The system according to any of the preceding embodiments wherein the analyzing component is further configured to measure an influence of connection rail and sleeper.

5104. The system according to any of the preceding embodiments wherein the analyzing component is further configured to measure an influence of the weight of the train on the tracks.

5105. The system according to any of the preceding embodiments wherein the analyzing component is further configured to measure an influence of the wear of the wheels on the speed of the train.

Below, method embodiments will be discussed. These embodiments are abbreviated by the letter "M" followed by a number. Whenever reference is herein made to "method embodiments", these embodiments are meant.

Ml. A method for monitoring properties of at least one train, the method comprising collecting at least one sensor data at least one time via a least one sensor arranged on at least one railway component, determining at least one property of the at least one train based on the at least one sensor data, and generating at least one determined property finding. M2. The method according to the preceding embodiment, wherein the method further comprises inferring an estimation value of the at least one property of the at least one train, and generating at least one estimated value.

M3. The method according to any of the 2 preceding embodiments, wherein the method comprises calibrating a physical model of the at least one property, and generating at least one calibrated physical model.

M4. The method according to the preceding embodiment wherein the physical model comprises a first level model and/or a second level model and/or a third level model.

M5. The method according to the any of the preceding method embodiments, wherein calibrating of the physical model is based on the at least one determined property finding.

M6. The method according to any of the preceding method embodiments and with feature of embodiment M3, wherein calibrating of the physical model is based on the at least one estimated value.

M7. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein the at least one estimated value comprises a weight estimation.

M8. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein the at least one estimated value comprises an axle patterns estimation.

M9. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein the at least one estimated value comprises a train speed estimation.

MIO. The method according to any of the preceding method embodiments and with features of embodiment M2, wherein the estimated value comprises a wheel health estimation. Mil. The method according to any of the preceding method embodiments, wherein estimating an estimation value of the at least one property of the at least one train is based on at least one undedicated measurement.

M12. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein calibrating of the physical model comprising calibrating the physical model to at least one acceleration data.

M13. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein calibrating of the physical model comprising calibrating the physical model to at least one estimation of train weight.

M14. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein calibrating of the physical model comprising calibrating the physical model to at least one speed data.

M15. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein calibrating of the physical model comprising calibrating the physical model to at least one wheel health data.

M16. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein calibrating of the physical model comprising calibrating the physical model to at least one axle pattern data.

M17. The method according to any of the preceding method embodiments and with features of embodiment M3, wherein the method comprises monitoring at least one physical parameter of at least one asset based on the at least one calibrated physical model.

M18. The method according to the preceding embodiment, wherein the method comprises generating at least one corrective measure, and applying the at least one corrective measure to the asset.

M19. The method according to any of the 2 preceding embodiments, wherein the method comprises evaluating an effectiveness of the at least corrective measure.

M20. The method according to any of the preceding method embodiments, wherein the method comprises performing a continuous monitoring, wherein the continuous monitoring comprises providing a series of measurements of the at least one physical parameters, wherein the method comprises an initial measurement and at least one subsequent measurement, wherein the method comprises comparing initial measurement and the at least one subsequent measurement to generate at least one evolution status, wherein the at least one evolution status is based on health status of the asset before and after a corrective measure.

M21. The method according to the preceding embodiment, wherein the method comprises determining an outcome status of the at least one corrective measure.

M22. The method according to any of the 2 preceding embodiments, wherein the method comprises quantifying a degree of effectiveness of the at least one corrective measure.

M23. The method according to the 2 preceding embodiments, wherein the degree of effectiveness is quantifying based on the outcome status of the at least one corrective measure.

M24. The method according to the preceding embodiment, wherein the method comprises processing the at least one sensor data via at least one processing component to generate at least one processed sensor data.

M25. The method according to the any of preceding embodiments, wherein estimating the value of the at least one property of the at least one train of railway component is based on the at least one processed sensor data.

M26. The method according to any of the preceding method embodiments, wherein the method comprises generating at least one railway component health status hypothesis of at least one component of the railway network.

M27. The method according to the preceding embodiment, wherein generating at least one railway component health status hypothesis of at least one component of the railway network is based on at the at least one processed sensor data.

M28. The method according to any of the preceding method embodiments, wherein the method comprises forecasting at least one railway health status of at least one component of the railway network.

M29. The method according to the preceding embodiment, wherein the method comprises forecasting at least one railway health status of at least one component of the railway network based on the at least one railway component health hypothesis. M30. The method according to any of the preceding embodiments, wherein the method comprises correlating the at least one estimated value with at least one railway component feature of the at least one railway component of the railway network.

M31. The method according to the preceding embodiment, wherein the value of the at least one property of the at least one train of at least one railway component of the railway network comprises at least one of: rail pad damping, rail pad stiffness, rail clamp damping, rail clamp stiffness, rail internal damping, rail shearing module, rail shearing coefficient, rail Young's modulus, rail density, rail shear tension, sleeper internal damping, sleeper shearing module, sleeper shearing coefficient, sleeper Young's modulus, sleeper density, sleeper shear tension, track bed damping, track bed stiffness, ballast shearing module, ballast shearing coefficient, ballast density, ballast cross-sectional area, ballast area moment of inertia.

M32. The method according to any of the preceding embodiments, wherein the method further comprises connecting the at least one sensor component to at least one server.

M33. The method according to any of the preceding method embodiments, wherein the method comprises retrieving at least one signal comprising at least one of the at least one sensor data, and the at least one processed sensor data.

M34. The method according to the preceding embodiment, wherein the at least one signal comprises a signal frequency content and a sampling rate, wherein the signal frequency is between 0 and 10,000 Hz, preferably between 50 and 8,000 Hz, more preferably between 50 and 5,000 Hz, and wherein the sampling rate is between 0 and 20 kHz, preferably between 0.1 and 8 kHz, more preferably between 1 and 5 kHz.

M35. The method according to any of the preceding method embodiments, wherein the method comprises unidirectionally connecting the at least one sensor component to the at least one server.

M36. The method according to any of the embodiments Ml to M29, wherein the method comprises bidirectionally connecting the at least one sensor component to the at least one server.

M37. The method according to any of the preceding embodiment, wherein the method comprises connecting the at least one server with the at least one processing component. M38. The method according to any of the preceding embodiment, wherein at least one of the at least one server comprises at least partially one of the at least one processing component.

M39. The method according to any of the preceding method embodiments, wherein the method comprises the at least one sensor component to supplying the at least one sensor data to the at least one sensor processing component.

M40. The method according to any of the preceding method embodiments, wherein the at least one processing component comprises a storing component, wherein the method comprising storing at least one of the at least one sensor data, and the at least one processed sensor data.

M41. The method according to any of the preceding method embodiments, wherein the sensor processing component comprises a sensor storing component configured to store at least one of the at least one sensor data, and the at least one processed sensor data.

M42. The method according to any of the preceding method embodiments and with features of embodiment M33, wherein the at least one signal comprises at least one of: displacement data, velocity data, and acceleration data.

M43. The method according to any of the preceding method embodiments, wherein the method comprises automatically transmitting to the at least one processing component the at least one sensor data.

M44. The method according to any of the preceding method embodiments, wherein the method comprises pre-processing the at least one sensor data via the at least one processing component.

M45. The method according to any of the preceding method embodiments, wherein the method further comprises supplying the at least one processing component with at least one neural network component. M46. The method according to any of the preceding method embodiments, wherein the method comprises feeding into the at least one neural network component the at least one sensor data.

M47. The method according to any of the preceding method embodiments and features of embodiment Ml, wherein the method comprises feeding into the at least one neural network component the at least one pre-processed sensor data.

M48. The method according to any of the preceding method embodiments and with features of embodiment M47, wherein the at least one neural network comprises at least one convolutional neural network layer.

M49. The method according to any of the preceding method embodiments and the features of embodiment M44, wherein the method comprises using the outcome of the preprocessing for predicting at least one trend of the at least one property of the at least one train.

M50. The method according to the preceding embodiment, wherein the method comprises predicting at least one trend of the at least one railway component based on predicting the at least one trend of the at least one property of the at least one train.

M51. The method according to any of the preceding method embodiments, wherein the method comprises teaching the at least one neural network the at least one property of the at least one train using at least one training database.

M52. The method according to any of the preceding method embodiments, wherein the at least one railway component comprises track bed.

M53. The method according to any of the preceding method embodiments, wherein the at least one railway component comprises at least one rail pad.

M54. The method according to any of the preceding method embodiments, wherein the at least one railway component comprises at least one sleeper.

M55. The method according to any of the preceding method embodiments, wherein the at least one railway component comprises at least one rail. M56. The method according to any of the preceding method embodiments, wherein the method further comprises generating at least one suggestion procedure to be applied on the at least one railway component.

M57. The method according to any of the preceding method embodiments, wherein the method further comprises generating at least one suggestion procedure to be applied on the at least one train.

M58. The method according to any of the preceding method embodiments, wherein the method comprises generating at least one action instruction to be applied on the at least one railway component.

M59. The method according to any of the preceding method embodiments, wherein the method comprises generating at least one action instruction to be applied on the at least one train.

M60. The method according to any of the 4 preceding embodiments, wherein the method further comprises prompting a user to implement at least one of the at least one suggestion procedure, and the at least one action instruction.

M61. The method according to any of the preceding method embodiments, wherein the at least one sensor data comprises a first sensor data and second sensor data.

M62. The method according to the preceding embodiment, wherein the first sensor data is different from the second sensor data.

M63. The method according to any of the preceding method embodiments, wherein the at least one time comprises a first time and a second time.

M64. The method according to the preceding embodiment, wherein the first time is different from the second time.

M65. The method according to any of the preceding method embodiments, wherein the method comprises estimating a specific physical property of the at least one railway component to generate a first physical property finding, correlating the first physical property finding to at least health status, and generating at least one final physical property finding comprising the health status of the at least one railway component.

M66. The method according to the preceding embodiment, wherein the estimating of the specific physical property of the at least one railway component is based on the at least one estimated value.

M67. The method according to the preceding embodiment and with features of any the embodiment M57, wherein the at least one suggestion procedure is based on the at least one final physical property finding.

M68. The method according to the preceding embodiment and with features of any the embodiments M57 or M58, wherein the at least one suggestion procedure and/or the at least one action instruction is based on the at least one final physical property finding.

M69. The method according to the preceding embodiment and with features of any the embodiments M57 or M58, wherein the at least one suggestion procedure is based on the at least one estimated value.

M70. The method according to the preceding embodiment, wherein and/or the at least one action instruction is based on the at least one estimated value.

M71. The method according to any of the preceding method embodiments, wherein the method comprises of generating at least one causal hypothesis upon outputting the at least one estimated value.

M72. The method according to the preceding embodiment, wherein the method comprises predicting at least one causal event for a divergence of the at least one estimated value.

M73. The method according to any of the preceding method embodiments, wherein the step of forecasting at least one railway health status of the at least one component of the railway network is based on at least one analytical approach.

M74. The method according to any of the preceding method embodiments, wherein the method comprises using at least one supervised learning method.

M75. The method according to any of the preceding method embodiments, wherein the method comprises using at least one unsupervised learning method. M76. The method according to any of the preceding method embodiments, wherein the method comprises automatically generating at least one interpreted dataset.

M77. The method according to any of the preceding method embodiments, wherein the method further comprises automatically generating at least one simulated sensor data.

M78. The method according to the preceding embodiment, wherein the method comprises calculating characteristics of the at least one simulated sensor data.

M79. The method according to any of the preceding method embodiments, wherein the method comprises at least one analytical approach.

M80. The method according to the preceding embodiment, wherein the at least one analytical approach comprises at least one of: signal filter processing, pattern recognition, probabilistic modeling, Bayesian methods, machine learning, supervised learning, unsupervised learning, reinforcement learning, statistical analytics, statistical models, principle component analysis, independent component analysis (ICA), dynamic time warping, maximum likelihood estimates, modeling, estimating, neural network, convolutional network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithms, particle filters, variations of Kalman filters, Markov models, and/or hidden Markov models.

M81. The method according to any of the preceding method embodiments, wherein the method comprises determining a safety threshold of the at least one train.

M82. The method according to the preceding embodiment, wherein the safety threshold comprises a specified class of the at least one train.

M83. The method according to the preceding embodiment, wherein the specified class data of the least one train comprises at least one of: a speed class data, and an axle weight class data.

M84. The method according to the preceding embodiment, wherein comprises measuring at least one of the speed class data and the axle weight class data.

M85. The method according to any of the preceding 4 embodiments, wherein the method further comprises monitoring the specified class data of the least one train, and adjusting at least one parameter of the specified class data.

M86. The method according to the preceding embodiment, wherein adjusting the at least one parameter comprises applying a filter.

M87. The method according to any of the preceding method embodiments, wherein estimating the at least one physical parameter comprises implementing at least one optimization step comprising at least one of: Newton methods, Levenberg-Marquardt, or other local optimization techniques, as well as variations genetic algorithms, particle swarm optimization, differential evolution, and simulated annealing.

M88. The method according to any of the preceding embodiments wherein the method comprises the step of automatically labelling the sensor data based on an already existing knowledgebase.

M89. The method according to any of the preceding method embodiments, wherein the method further comprises determining at least one mechanical property of a subsoil.

M90. The method according to the preceding embodiment, wherein the at least one mechanical property of the subsoil comprises at least one of stiffness and dampening.

M91. The method according to any of the preceding method embodiments, wherein the method further comprises determining at least one mechanical property of a ballast.

M92. The method according to the preceding embodiment, wherein the at least one mechanical property of the ballast comprises at least one of stiffness, dampening, and a presence of voids under sleepers.

M93. The method according to any of the preceding method embodiments, wherein the method further comprises determining at least one physical properties of at least one track.

M94. The method according to the preceding embodiment, wherein the at least one physical property of the at least one track comprises at least one of stiffness, dampening, behavior-relevant indicators of surface geometry, and looseness of clamps.

M95. The method according to any of the preceding method embodiments, wherein the method further comprises determining a response to temperature of the at least one estimated physical parameter to temperature. M96. The method according to any of the preceding method embodiments, wherein the method further comprises determining geometry of a whole track.

M97. The method according to any of the preceding method embodiments, wherein the method further comprises determining status of a least one frog.

M98. The method according to the preceding embodiment, wherein the status comprises at least one of dip angle and surface defects.

M99. The method according to any of the preceding method embodiments, wherein the method further comprises determining at least one property of train travelling over at least one monitored track.

M100. The method according to the preceding embodiment, the at least one property comprising at least one of health status of wheels, speed and weight at each axle.

M101. The method according to any of the preceding method embodiments, wherein the method further comprises determining a safety margin of track for: a particular train weight and/or train speed.

M102. The method according to the preceding embodiment, wherein the method comprises determining the safety margin based on the at least one estimated physical property.

M103. The method according to any of the preceding method embodiments, wherein the method further comprises monitoring at least one physical property of the at least one track.

M104. The method according to any of the preceding method embodiments, wherein the method further comprises forecasting an effect of traffic of different classes of trains on the at least one physical parameter.

M105. The method according to any of the preceding method embodiments, wherein the method further comprises generating at least one of a timing recommendation, and a type of maintenance action, for improving at least one specified physical parameter of the at least one physical parameter. M106. The method according to any of the preceding method embodiments, wherein the method further comprises generating at least one of a speed restrictions recommendation.

M107. The method according to any of the preceding method embodiments, wherein the method further comprises assigning a proportionality of damage to the at least one physical properties, wherein proportionality damage is assigned to particular train class.

M108. The method according to any of the preceding method embodiments, wherein the method comprises carrying out the method according to any of the preceding embodiments via the system according to any of the preceding system embodiments.

M109. The method according to any of the preceding method embodiments wherein method comprises calculating the estimated value based on a risk estimation value, such as risk associated with a hazard.

Ml 10. The method according to any of the preceding method embodiments comprising building the simulation model configured with a vehicle-track interaction in a multi-body simulation environment.

Mill. The method according to any of the preceding method embodiments wherein the method comprises executing the simulation model to extract at least sleeper-ballast contact pressure.

Ml 12. The method according to any of the preceding method embodiments wherein the method comprises executing the simulation model to extract at least sleeper acceleration.

Ml 13. The method according to any of the preceding method embodiments comprising building the simulation model based on a track settlement model.

M114. The method according to the preceding method embodiment wherein simulation model is based on time series analysis technique assuming that the ballast is uncompressed at time = 0 sec.

M115. The method according to any of the preceding method embodiment wherein the method comprises calculating at least permanent deformation in at least ballast geometry based on the track settlement model. Ml 16. The method according to any of the preceding method embodiments wherein the method comprises generating a rail pad model, wherein the rail pad model is used to measure a static stiffness of the rail pads as a function of time.

Ml 17. The method according to any of the preceding method embodiments wherein the method comprises measuring dynamic stiffness and/or damping of the rail pad using the rail pad model.

Ml 18. The method according to any of the preceding method embodiments wherein the method comprises measuring an influence of the train speed on fatigue criterion of the railway tracks.

S106. The system according to any of the preceding system embodiments, wherein the system is configured to perform the method according to any of the preceding method embodiments.

Below, use embodiments will be discussed. These embodiments are abbreviated by the letter "U" followed by a number. Whenever reference is herein made to "use embodiments", these embodiments are meant.

Ul. Use of the system according to any of the preceding embodiments, for carrying out the method according to any of the preceding method embodiments.

U2. Use of the method according to any of the preceding method embodiments and the system according to any of the preceding embodiments, for analyzing a property of at least one train.

U3. Use of the method according to any of the preceding method embodiments and the system according to any of the preceding embodiments, for analyzing a property of at least railway infrastructure.

Below, program embodiments will be discussed. These embodiments are abbreviated by the letter "C" followed by a number. Whenever reference is herein made to "program embodiments", these embodiments are meant.

Cl. A computer-implemented program comprising instructions which, when executed by a user-device, causes the user-device to carry out the method steps according to any of the preceding method embodiments. C2. A computer-implemented program comprising instructions which, when executed by a server, causes the at least one server to carry out the method steps according to any of the preceding method embodiments.

C3. A computer-implemented program comprising instructions which, when executed causes by a user-device, causes the user-device and a server to carry out the method steps according to any of the preceding method embodiments.

Brief description of the drawings

The present invention will now be described with reference to the accompanying drawings, which illustrate embodiments of the invention. These embodiments should only exemplify, but not limit, the present invention.

Fig. 1 depicts a schematic representation of a railway network and system arranged at the railway network;

Fig. 2 depicts a system for monitoring a railway network according to embodiments of the present invention;

Fig. 3 depicts a schematic of a computing device; and

Fig. 4 depicts an exemplary embodiment of a system configured to execute multilevel modeling.

Detailed description of the drawings

In the following description, a series of features and/or steps are described. The skilled person will appreciate that unless explicitly required and/or unless requires by the context, the order of features and steps is not critical for the resulting configuration and its effect. Further, it will be apparent to the skilled person that irrespective of the order of features and steps, the presence or absence of time delay between steps can be present between some or all of the described steps.

It is noted that not all the drawings carry all the reference signs. Instead, in some of the drawings, some of the reference signs have been omitted for sake of brevity and simplicity of illustration. Embodiments of the present invention will now be described with reference to the accompanying drawings. Fig. 1 depicts a schematic representation of a railway network and system arranged at the railway network. In simple terms, the system may comprise a railway section with the railway 1 itself, comprising rails 10 and sleepers 3. Instead of the sleepers 3 also a solid bed for the rails 10 can be provided.

Moreover, a further example of constitutional elements is conceptually represented a mast, conceptually identified by reference numeral 6. Such constitutional elements are usually arranged at or in the vicinity of railways. Furthermore, a tunnel is shown, conceptually identified by reference numeral 5. It should be understood that other constructions, buildings etc. may be present and also used for the present invention as described before and below.

For instance, a first sensor 2 can be arranged on one or more of the sleepers. The sensor 2 can be an acceleration sensor and/or any other kind of railway specific sensor. Examples have been mentioned before.

Further, a second sensor 9 can also arranged on another sleeper distant from the first sensor 2. Although it seems just a small distance in the present example, those distances can range from the distance to the neighboring sleeper to one or more kilometers. Other sensors can be used for attachment to the sleepers as well. The sensors can further be of different kind - such as where the first sensor 2 may be an acceleration sensor, the second sensor 9 can be a magnetic sensor or any other combination suitable for the specific need. The variety of sensors are enumerated before.

In one embodiment, any of the sensors, for example, the first sensor 2 and/or the second sensor 9, can directly be attached to the rail.

The sensors, for example the first sensor 2 and/or the second sensor 9, further comprise a wireless sensor network. The sensor node can transmit data to a base station (not shown here). The base station can be installed to the railway component. The base station can also be installed in the surroundings of the railway component. The base station can also be a remote base station. The communication module between the base station and the sensor node (s) can comprise, for example Xbee with a frequency of 868 MHz, but is not limited to this.

The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can also be installed in cases and inserted inside the railway component, for example inside a special hole carved into the concrete. The case can also be attached to the railway component using fixers. The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can be obtaining at least one sensor data based on acceleration, inclination, distance, etc.

The sensor node (s), for example the first sensor 2 and/or the second sensor 9, may further be divided into group, for example based on the distance. The sensor node (s), for example the first sensor 2 and/or the second sensor 9 lying within a pre-determined distance may be controlled by one base station. The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can also be installed on the moving railway component such as on-board of a vehicle. The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can comprise an amplifier to amplify any signal received by the base station.

The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can be installed such that the sensor node lying within one group can communicate with their base station in one-hop. The base station can receive information from its 'neighbors' and retransmit all the information to the server 800.

The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can comprise sensor(s). The sensor can be accelerometers, such as Sensor4PRI for example ADCL 345, SQ-SVS etc. The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can comprise inclinometers, such as SQ-SI-360DA, SCA100T-D2, ADXL345 etc.

The sensor node can further comprise distance sensors. The distance sensors can be configured to at least measure the distance between slab tracks, using infrared and/or ultrasonic. The distance sensor can be for example, MB1043, SRF08, PING, etc.

The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can comprise visual sensors, such as 3D cameras, speed enforcement cameras, traffic enforcement cameras, etc. It may be noted that sensor node(s) may comprise sensors to observe the physical environment of the infrastructure the sensor node(s) are installed in. For example, but not limited to, temperature sensor, humidity sensor, altitude sensor, pressure sensor, GPS sensor, water pressure sensor, piezometer, multi-depth deflectometers (MDD), acceleration.

The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can be installed according to a protocol based on routing trees to be able to transmit information to the base station. Once the information has been received, a cellular network can be used to send at least one sensor data to a remote server 800. The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can comprise an analog-to-digital converter, a micro controller, a transceiver, power and memory. One or more sensor(s) can be embedded in different elements and can be mounted on boards to be attached to the railway component. The sensor node (s), for example the first sensor 2 and/or the second sensor 9, can also comprise materializing strain gauges, displacement transducers, accelerometers, inclinometers, acoustic emission, thermal detectors, among others. The analog signal outputs generated by the sensors can be converted to digital signals that can be processed by digital electronics. The data can then be transmitted to the base station by a microcontroller through a radio transceiver. All devices can be electric or electronic components supported by power supply, which can be provided through batteries or by local energy generation (such as solar panels), the latter mandatory at locations far away from energy supplies.

The at least one sensor data collected from the sensor node (s), for example the first sensor 2 and/or the second sensor 9, can be transferred to the base station using wireless communication technology such as Wi-Fi, -Bluetooth, ZigBee or any other proprietary radio technologies suitable for the purpose. For example, the ZigBee network can be advantageous to consumes less power. On the other hand, for transmitting the input data from the base station to the server 800 long-range communication such a cellular network or satellite can be used as well as wired technologies based on optical fiber.

Due to the short transmission range, communications from sensor nodes may not reach the base station, a problem that can be overcome by adopting relay nodes to pass the data from the sensor node (s), for example the first sensor 2 and/or the second sensor 9.

Another sensor 7, which may be different or the same kind of sensor, can be attached, for example, to the mast 6 or any other structure. This may be a different kind of sensor, such as, for example, an optical, temperature, even acceleration sensor, etc. A further kind of sensor, for example sensor 8, can be arranged above the railway as at the beginning or within the tunnel 5. This could, for example, be a height sensor for determining the height of a train, an optical sensor, a doppler sensor etc. It will be understood that all those sensors mentioned here and/or before are just non-limiting examples.

Furthermore, the sensors can be configured to submit the at least one sensor data via a communication network, such as a wireless communication network. As the communication network bears several advantages and disadvantages regarding availability, transmittal distance, costs etc. the transmittal of at least one sensor data is optimized as described herein before and below. Fig. 2 depicts a system 100 monitoring a railway network. In simple terms, the system 100 may comprise a sensor component 200, a processing component 300, a storing component 400, an analyzing component 500 and a server 600.

In one embodiment, the sensor component 200 may comprise a plurality of sensor units, and each may comprise a plurality of sensor nodes. Therefore, the sensor component 200 may also be referred to as a plurality of sensor components 200.

Additionally, or alternatively, the sensor component may be configured to sample information relevant to a railway network, for instance, electric current based information of a given component and/part of a railway network.

In one embodiment, the processing 300 component may comprise a standalone component configure to retrieve information from the sensor 200. Additionally, or alternatively, the processing component may be configured to bidirectionally communicate the storing component 300 and the analyzing component 500. For instance, the processing component 300 may transfer raw at least one sensor data to the storing component 400, wherein the raw at least one sensor data may be stored until the processing component 300 may require said data for processing to generate a processed at least one sensor data. In another embodiment, the processing component 300 may also transfer processed at least one sensor data to the storing component 400. In a further embodiment, the processing component may also retrieve data from the storing component 400.

In one embodiment, the analyzing component 500 may be configured to bidirectionally communicate with the processing component 300, the storing component 400 and/or the server 600. It will be understood that the communication of the analyzing component 500 with the other components may take place independent and/or simultaneously one from another.

In one embodiment, the processing component 300 may also be integrated with at least one of the sensors 200. In other words, the processing component 300 may also comprise an imbedded module of the sensors 200.

In embodiment, the analyzing component 500 may be configured to process at least one sensor data based on at least one analytical approach, each approach comprising at least one of signal filter processing, pattern recognition, probabilistic modeling, Bayesian schemes, machine learning, supervised learning, unsupervised learning, reinforcement learning, statistical analytics, statistical models, principle component analysis, independent component analysis (ICA), dynamic time warping, maximum likelihood estimates, modeling, estimating, neural network, convolutional network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithms, Markov models, and/or hidden Markov models.

The server 600 may comprise one or more modules configured to receive information from the analyzing component 500.

In another embodiment of the presentation invention, the sensor 200, the processing component 300, the storing component 400 and the analyzing component may comprise an integrated module configured to execute subsequently the tasks corresponding to each individual component, and transfer a final processed analyzed at least one sensor data to the server 600. In simple words, in one embodiment the sensor 200, the processing component 300, the storing component 400 and the analyzing component 500 may comprises modules of a single component.

In one embodiment, the server 600 may retrieve information from the analyzing component 500, and further may provide information to the analyzing component 500, for example, operation parameters. It will be understood that each component may receive a plurality of operation parameters, for instance, the processing component 300 may be commanded to execute a preprocessing of the data received from the sensors 200.

Alternatively or additionally, the processing component 300 may be instructed to transmit the original data received from the sensors 200, i.e., the data coming from the sensors 200 can be transferred directly to the next component without executing any further task. It will be understood that the component may also be configured to perform a plurality of tasks at the same time, e.g., processing the data coming from the sensor 200 before transferring to the next component and transferring the data coming from the sensors 200 without any processing.

In one embodiment, the server 600 may comprise a cloud server, a remote server and/or a collection of different type of servers. Therefore, the server 600 may also be referred to as cloud server 600, remote server 600, or simple as servers 500. In another embodiment, the servers 500 may also converge in a central server.

It will be understood that the server 600 may also be in bidirectional communication with the storing component 400, the processing component or the sensor component 200 without passing through the analyzing component 500 or any other intermediate component. For this purpose, each component may also comprise a remote communication unit configured to establish a remote communication between a component, e.g., sensor component 200, with the server 600.

The storing component 400 may be configured to receive information from the server 600 for storage. In simple words, the storing component 400 may store information provided by the servers 600. The information provided by the server 600 may include, for example, but not limited to, data obtained by sensors 200, data processed by the processing component 500 and any additional data generated in the servers 600. It will be understood that the servers 600 may be granted access to the storing component 400 comprising, inter alia, the following permissions, reading the data allocated in the storing component 400, writing and overwriting the data stored in the storing component 400, control and modify the storage logic and the data distribution within the storing component 400.

In one embodiment of the present invention the server 600 may be configured transmit a signal to other component of the railway system based upon health status information retrieved from sensors 200. For instance, a giving health status data is provided by the server 600 and subsequently the server 600 generates a signal containing instructions, which are transmitted to the railway system for implementation. The set of instructions may comprise, inter alia, generating a hypothesis as regards the health status of the railway network and/or a failure hypothesis, which may comprise instructions to be implemented before a failure occurs on the railway network, such as switching rolling unit from on track to another. Furthermore, the signal may be based on at least one analytical approach, each approach comprising at least one of signal filter processing, pattern recognition, probabilistic modeling, Bayesian schemes, machine learning, supervised learning, unsupervised learning, reinforcement learning, statistical analytics, statistical models, principle component analysis, independent component analysis (ICA), dynamic time warping, maximum likelihood estimates, modeling, estimating, neural network, convolutional network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithms, Markov models, and/or hidden Markov models.

In one embodiment, the sensors 200 may, inter alia, adopt a configuration that allows identifying trains, their speeds and their wear effect on the tracks. The data gathered by the sensors 200 may constitute the basis for the server 600 to generate instructions for the activation of the switches. In simple words, if a train is approaching this part of the network, the sensors 200 may retrieve data that may allow activating the switches in order to redirect the trains, for example, from track 1 to track 2, according to their speed and/or wear effect. The data gathered by the sensors 200 may be communicated to the server 600, which may subsequently transmit the information and the corresponding instructions to the nearest assets, for example, the nearest switch, which may consequently be activated to control the traffic on the tracks. Furthermore, in one embodiment of the present invention, the system 100 may estimate the health status of components of the railway network and may further generate a health status and/or failure hypothesis that may allow to forecast the suitability of the component of the railway network to allocate rolling units. Such hypothesis may be based on at least one analytical approach, each approach comprising at least one of signal filter processing, pattern recognition, probabilistic modeling, Bayesian schemes, machine learning, supervised learning, unsupervised learning, reinforcement learning, statistical analytics, statistical models, principle component analysis, independent component analysis (ICA), dynamic time warping, maximum likelihood estimates, modeling, estimating, neural network, convolutional network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithms, Markov models, and/or hidden Markov models.

In another embodiment of the present invention, the system 100 may determine that a particular part and/or component of the railway network, for instance, a given section of track and/or a switch, is required to be replaced and/or maintain before a given date to avoid failure of the railway.

In one embodiment of the present invention, the system 100 may also determine that a particular rolling stock may pass through a component or portion of the railway network requiring maintenance, reparation or replacement, however, due to work schedule it may be prompt to failure if an inadequate rolling unit passes through. This approach may be advantageous, as it may allow to reduce failure of railway networks, which may be achieved by monitoring, evaluating and forecasting optimal operation conditions of the railway network.

Furthermore, the system 100 may be configured to predict a future status of the railway network and based on that may determine an optimal operation conditions using data analysis based on at least one analytical approach, each approach comprising at least one of signal filter processing, pattern recognition, probabilistic modeling, Bayesian schemes, machine learning, supervised learning, unsupervised learning, reinforcement learning, statistical analytics, statistical models, principle component analysis, independent component analysis (ICA), dynamic time warping, maximum likelihood estimates, modeling, estimating, neural network, convolutional network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithms, Markov models, and/or hidden Markov models.

In more simple words, determinations of the system 100 may directly be used forecast xxx failure, which may be advantageous for planning and execution of maintenance and/or inspections of railway network, which may further allow to minimize downtime of single machines and more importantly an adjacent railway network. Such monitoring, analyzing and forecasting may be based on machine learning comprising predicting health status hypothesis and/or failure hypothesis based on at least one analytical approach, each approach comprising at least one of signal filter processing, pattern recognition, probabilistic modeling, Bayesian schemes, machine learning, supervised learning, unsupervised learning, reinforcement learning, statistical analytics, statistical models, principle component analysis, independent component analysis (ICA), dynamic time warping, maximum likelihood estimates, modeling, estimating, neural network, convolutional network, deep convolutional network, deep learning, ultra-deep learning, genetic algorithms, Markov models, and/or hidden Markov models.

Fig. 3 depicts a schematic of a computing device 1000. The computing device 1000 may comprise a computing unit 35, a first data storage unit 30A, a second data storage unit 30B and a third data storage unit 30C.

The computing device 1000 can be a single computing device or an assembly of computing devices. The computing device 1000 can be locally arranged or remotely, such as a cloud solution.

On the different data storage units 30 the different data can be stored, such as the genetic data on the first data storage 30A, the time stamped data and/or event code data and/or phenotypic data on the second data storage 30B and privacy sensitive data, such as the connection of the before-mentioned data to an individual, on the thirds data storage 30C.

Additional data storage can be also provided and/or the ones mentioned before can be combined at least in part. Another data storage (not shown) can comprise data specifying for instance, air temperature, rail temperature, position of blades, model of xxx, position of xxx and/or further railway network related information. This data can also be provided on one or more of the before-mentioned data storages.

The computing unit 35 can access the first data storage unit 30A, the second data storage unit 30B and the third data storage unit 30C through the internal communication channel 160, which can comprise a bus connection 160.

The computing unit 30 may be single processor or a plurality of processors, and may be, but not limited to, a CPU (central processing unit), GPU (graphical processing unit), DSP (digital signal processor), APU (accelerator processing unit), ASIC (application-specific integrated circuit), ASIP (application-specific instruction-set processor) or FPGA (field programable gate array). The first data storage unit 30A may be singular or plural, and may be, but not limited to, a volatile or non-volatile memory, such as a random-access memory (RAM), Dynamic RAM (DRAM), Synchronous Dynamic RAM (SDRAM), static RAM (SRAM), Flash Memory, Magneto-resistive RAM (MRAM), Ferroelectric RAM (F-RAM), or Parameter RAM (P-RAM).

The second data storage unit 30B may be singular or plural, and may be, but not limited to, a volatile or non-volatile memory, such as a random-access memory (RAM), Dynamic RAM (DRAM), Synchronous Dynamic RAM (SDRAM), static RAM (SRAM), Flash Memory, Magneto-resistive RAM (MRAM), Ferroelectric RAM (F-RAM), or Parameter RAM (P-RAM). The third data storage unit 30C may be singular or plural, and may be, but not limited to, a volatile or non-volatile memory, such as a random-access memory (RAM), Dynamic RAM (DRAM), Synchronous Dynamic RAM (SDRAM), static RAM (SRAM), Flash Memory, Magneto-resistive RAM (MRAM), Ferroelectric RAM (F-RAM), or Parameter RAM (P-RAM).

It should be understood that generally, the first data storage unit 30A (also referred to as encryption key storage unit 30A), the second data storage unit 30B (also referred to as data share storage unit 30B), and the third data storage unit 30C (also referred to as decryption key storage unit 30C) can also be part of the same memory. That is, only one general data storage unit 30 per device may be provided, which may be configured to store the respective encryption key (such that the section of the data storage unit 30 storing the encryption key may be the encryption key storage unit 30A), the respective data element share (such that the section of the data storage unit 30 storing the data element share may be the data share storage unit 30B), and the respective decryption key (such that the section of the data storage unit 30 storing the decryption key may be the decryption key storage unit 30A).

In some embodiments, the third data storage unit 30C can be a secure memory device 30C, such as, a self-encrypted memory, hardware-based full disk encryption memory and the like which can automatically encrypt all of the stored data. The data can be decrypted from the storing component only upon successful authentication of the party requiring to access the third data storage unit 30C, wherein the party can be a user, computing device, processing unit and the like. In some embodiments, the third data storage unit 30C can only be connected to the computing unit 35 and the computing unit 35 can be configured to never output the data received from the third data storage unit 30C. This can ensure a secure storing and handling of the encryption key (i.e., private key) stored in the third data storage unit 30C. In some embodiments, the second data storage unit 30B may not be provided but instead the computing device 1000 can be configured to receive a corresponding encrypted share from the database 60. In some embodiments, the computing device 1000 may comprise the second data storage unit 30B and can be configured to receive a corresponding encrypted share from the database 60.

The computing device 1000 may comprise a further storing component 140 which may be singular or plural, and may be, but not limited to, a volatile or non-volatile memory, such as a random-access memory (RAM), Dynamic RAM (DRAM), Synchronous Dynamic RAM (SDRAM), static RAM (SRAM), Flash Memory, Magneto-resistive RAM (MRAM), Ferroelectric RAM (F-RAM), or Parameter RAM (P-RAM). The storing component 140 may also be connected with the other components of the computing device 1000 (such as the computing component 35) through the internal communication channel 160.

Further the computing device 1000 may comprise an external communication component 130. The external communication component 130 can be configured to facilitate sending and/or receiving data to/from an external device (e.g., backup device 10, recovery device 20, database 60). The external communication component 130 may comprise an antenna (e.g., WIFI antenna, NFC antenna, 2G/3G/4G/5G antenna and the like), USB port/plug, LAN port/plug, contact pads offering electrical connectivity and the like. The external communication component 130 can send and/or receive data based on a communication protocol which can comprise instructions for sending and/or receiving data. Said instructions can be stored in the storing component 140 and can be executed by the computing unit 35 and/or external communication component 130. The external communication component 130 can be connected to the internal communication component 160. Thus, data received by the external communication component 130 can be provided to the storing component 140, computing unit 35, first data storage unit 30A and/or second data storage unit 30B and/or third data storage unit 30C. Similarly, data stored on the storing component 140, first data storage unit 30A and/or second data storage unit 30B and/or third data storage unit 30C and/or data generated by the commuting unit 35 can be provided to the external communication component 130 for being transmitted to an external device.

In addition, the computing device 1000 may comprise an input user interface 110 which can allow the user of the computing device 1000 to provide at least one input (e.g., instruction) to the computing device 100. For example, the input user interface 110 may comprise a button, keyboard, trackpad, mouse, touchscreen, joystick and the like. Additionally, still, the computing device 1000 may comprise an output user interface 120 which can allow the computing device 1000 to provide indications to the user. For example, the output user interface 110 may be a LED, a display, a speaker and the like.

The output and the input user interface 100 may also be connected through the internal communication component 160 with the internal component of the device 100.

The processor may be singular or plural, and may be, but not limited to, a CPU, GPU, DSP, APU, or FPGA. The memory may be singular or plural, and may be, but not limited to, being volatile or non-volatile, such an SDRAM, DRAM, SRAM, Flash Memory, MRAM, F-RAM, or P- RAM.

The data processing device can comprise means of data processing, such as, processor units, hardware accelerators and/or microcontrollers. The data processing device 20 can comprise storing components, such as, main memory (e.g., RAM), cache memory (e.g., SRAM) and/or secondary memory (e.g., HDD, SDD). The data processing device can comprise busses configured to facilitate data exchange between components of the data processing device, such as, the communication between the storing components and the processing components. The data processing device can comprise network interface cards that can be configured to connect the data processing device to a network, such as, to the Internet. The data processing device can comprise user interfaces, such as:

■ output user interface, such as: o screens or monitors configured to display visual data (e.g., displaying graphical user interfaces of railway network status), o speakers configured to communicate audio data (e.g., playing audio data to the user),

■ input user interface, such as: o camera configured to capture visual data (e.g., capturing images and/or videos of the user), o microphone configured to capture audio data (e.g., recording audio from the user), o keyboard configured to allow the insertion of text and/or other keyboard commands (e.g., allowing the user to enter text data and/or other keyboard commands by having the user type on the keyboard) and/or trackpad, mouse, touchscreen, joystick - configured to facilitate the navigation through different graphical user interfaces of the questionnaire. The data processing device can be a processing unit configured to carry out instructions of a program. The data processing device can be a system-on-chip comprising processing units, storing components and busses. The data processing device can be a personal computer, a laptop, a pocket computer, a smartphone, a tablet computer. The data processing device can be a server, either local and/or remote. The data processing device can be a processing unit or a system-on-chip that can be interfaced with a personal computer, a laptop, a pocket computer, a smartphone, a tablet computer and/or user interface (such as the upper-mentioned user interfaces).

Fig. 4 depicts an exemplary embodiment of a system configured to execute a multilevel modeling. The system (100) and/or the analyzing component (500) can be configured to generate a first level model (Level 1). The first level model (Level 1) can be based on the at least one estimation value. The estimation value can be a representation of a feature associated with railway infrastructure. The system can be configured to estimate the at least one estimation value of the at least one train, preferably based on the frequency. The system can further be configured to estimate the at least one estimation value of at least one property of the at least one train. The system can further be configured to estimate the at least one estimation value of the railway infrastructure. It should be noted that the property can comprise a geometric property, locomotive property, kinematic property, physical property, etc. The analyzing component (500) can be configured to at least partially determine the at least one estimation value of the at least one train. In general estimation value can further comprises processed sensor data. The multilevel modeling as shown in this exemplary embodiment can be used when the sensor data is nested data.

The first level model (Level 1) can also be based on at least one estimated value (20, 21...29). The estimated value (20, 21..29) can be an inferred estimation value. The estimated value (20, 21...29) can comprise, for example, a weight estimation of at least one component in the railway infrastructure. In such embodiments the railway infrastructure can comprise the train, such as weight estimation may comprise weight estimation of the train. The estimated value (20, 21...29) can further comprise an axle patterns estimation or the like. The estimated value (20, 21...29) and/or the estimation value can comprise at least one physical, chemical and/or environmental parameter in railway infrastructure. The first level model (Level 1) can be configured to provide a weighted average of pooled model estimates. The multilevel modeling can further comprise Bayesian hierarchical modeling. This modeling can particularly be advantageous when the at least two estimated values (20, 21... ,29) are independent of each other. The estimated values (20, 21...29) can be nested within one another at different levels (Level 1, 2, 3) as shown in this exemplary embodiment. The system can further be configured to generate at least a first interaction score (40, 41). The first interaction score can comprise a coefficient, such as a regression coefficient associating the at least two estimated values (20, 21...29). For example, the sensor data can comprise acceleration data (20, 21...29) of a train, in some embodiments a higher acceleration can imply a void in the rocks, in such embodiment the model may generate the first interaction score (40) associating higher acceleration (20) to the void (23). The first interaction score (40, 41) can comprise a random coefficient, random parameter model, nested data models and/or split plot designs. The system can be configured to generate a second level model (Level 2). The second level model (Level 2) can be configured to be based on the at least first interaction score (41, 42).

The second level model (Level 2) can further be configured to generate a second interaction score (60). The second interaction score (60) can be based on the variable (50, 51, 54). The variable (50, 51, 54) can comprise at least one regression coefficient. The variable (50, 51, 54) can further comprise the error component. In some embodiments the variable (50, 51, 54) can further comprise a hyperprior. For example, a second level model (Level 2) can comprise the second interaction score (40, 41) associating the rail-wheel contact force with the weight of the train. And based on this may calculate the dip angle on the tracks.

The system can further generate a third level model (Level 3), wherein the third level model (Level 3) can be based on the second interaction score (60). The third level model (Level 3) may indicate a position on the tracks where the acceleration is increased. It is to be noted that the term railway infrastructure can also comprise the vehicle in the railway structure, such as the train. In general, where ever the term train properties are mentioned this can comprise properties of the vehicle in a railway infrastructure, such as speed of the train, axel pattern, category of the train (passenger train, cargo, etc.), weight of the train, properties related to wheels of the train (flat and/or deformed), etc. Additionally, and/or optionally, the term railway infrastructure can comprise geometric and/or physical properties of the infrastructure, such as stiffness and dampening of subsoil, ballast or rail pads, presence of void under sleeper, eigenfrequencies of sleepers, presence of surface defects on rail, dip angle of frig, geometry etc.

While in the above, a preferred embodiment has been described with reference to the accompanying drawings, the skilled person will understand that this embodiment was provided for illustrative purpose only and should by no means be construed to limit the scope of the present invention, which is defined by the claims. Reference numbers and letters appearing between parentheses in the claims, identifying features described in the embodiments and illustrated in the accompanying drawings, are provided as an aid to the reader as an exemplification of the matter claimed. The inclusion of such reference numbers and letters is not to be interpreted as placing any limitations on the scope of the claims.

The term "at least one of a first option and a second option" is intended to mean the first option or the second option or the first option and the second option.

Whenever a relative term, such as "about", "substantially" or "approximately" is used in this specification, such a term should also be construed to also include the exact term. That is, e.g., "substantially straight" should be construed to also include "(exactly) straight".

Whenever steps were recited in the above or also in the appended claims, it should be noted that the order in which the steps are recited in this text may be accidental. That is, unless otherwise specified or unless clear to the skilled person, the order in which steps are recited may be accidental. That is, when the present document states, e.g., that a method comprises steps (A) and (B), this does not necessarily mean that step (A) precedes step (B), but it is also possible that step (A) is performed (at least partly) simultaneously with step (B) or that step (B) precedes step (A). Furthermore, when a step (X) is said to precede another step (Z), this does not imply that there is no step between steps (X) and (Z). That is, step (X) preceding step (Z) encompasses the situation that step (X) is performed directly before step (Z), but also the situation that (X) is performed before one or more steps (Yl), ..., followed by step (Z). Corresponding considerations apply when terms like "after" or "before" are used.