LINE OF A TRANSPORTATION NETWORK

DESCRIPTION :

The present invention relates to an apparatus and a method for monitoring an overhead contact line of a transportation network, in particular an overhead contact line for trains , trams , trolley buses , or the like .

As is known, overhead line equipment ( OLE ) is very complex, especially that which is employed on high-speed railway lines , because , in order to trans fer the necessary power, said lines must be able to withstand a rather high contact force ( of the order of at least several hundreds of Newtons ) , which, due to the speed of the train, is exerted for a short time only, thus generating oscillations that need to be appropriately damped . For this reason, overhead line equipment includes a large number of components , so that it is quite likely that failures may occur somewhat frequently .

Therefore , overhead line equipment requires frequent inspection, both visual and by means of dedicated measurement instruments , on at least a monthly basis ; moreover, such inspection frequency is also determined according to the type of line and the traf fic thereon . For this purpose , a camera is used which is positioned on the roof of an electric locomotive to frame the electric line , the pantograph, and their point of contact ; said camera is configured for shooting a sequence of frames and/or acquiring single images , which are then saved, stored and, lastly, trans ferred to a control centre , where they will be subsequently viewed by one or more skilled technicians , who will then decide what kind of maintenance should be carried out in order to prevent the overhead line from failing, e . g . breaking and/or falling to the ground .

However, such an approach is unsatis factory in preventing faults , because several days may pass between the acquisition of the images and the remote inspection of the same . In some situations ( and especially on high-traf fic transportation lines ) , such a lapse of time is too long, since very often a breakdown or even an accident will have already occurred when the technician checks the images .

The present invention aims at solving these and other problems by providing an apparatus for monitoring an overhead contact line of a transportation network .

Furthermore , the present invention aims at solving these and other problems by providing also a method for monitoring an overhead contact line of a transportation network .

The basic idea of the present invention is to use a neural network for determining a monitoring datum which represents a state of the overhead contact line , wherein said neural network is trained in a manner such that , when inputting to said neural network at least one training image representing a second overhead contact line and a second pantograph in an abnormal operating condition, said neural network will output a second monitoring datum indicating the occurrence of an abnormal operating condition, and wherein said at least one monitoring datum is transmitted to a supervision computer via transmission means .

This makes it possible to determine the occurrence of abnormal operating conditions the very first time a train runs along the line , without requiring human intervention . It will thus be possible to immediately take the necessary steps ( e . g . to limit the maximum speed along a section of the line ) and/or to carry out preventive maintenance , so as to reduce the probability that faults or even accidents might occur .

It must be pointed out that the solution described herein can be used by any transportation or hauling means capable of circulating along the line ( e . g . a trolley, a vehicle , a railway carriage of a freight train, or the like ) or in the vicinity thereof ( e . g . remotely controlled vehicles or aircraft ) , without requiring the use of a dedicated means , e . g . a diagnostic train .

Further advantageous features of the present invention will be set out in the appended claims .

These features as well as further advantages of the present invention will become more apparent in the light of the following description of a preferred embodiment thereof as shown in the annexed drawings , which are provided herein merely by way of non-limiting example , wherein :

- Fig . 1 shows a railway monitoring system comprising an apparatus for monitoring an overhead contact line of a transportation network according to the invention;

- Fig . 2 shows a block diagram of the apparatus of Fig . 1 ;

- Fig . 3 shows a flow chart of a method for monitoring an overhead contact line of a transportation network according to the invention;

- Fig . 4 shows a set of training images , each one representing an abnormal operating condition;

- Fig . 5 shows a portion of an overhead contact line of a transportation network in a normal operating condition .

In this description, any reference to "an embodiment" will indicate that a particular configuration, structure or feature is comprised in at least one embodiment of the invention . Therefore , expressions such as " in an embodiment" and the like , which may be found in di f ferent parts of this description, will not necessarily refer to the same embodiment . Moreover, any particular configuration, structure or feature may be combined as deemed appropriate in one or more embodiments . The references below are therefore used only for simplicity' s sake , and shall not limit the protection scope or extension of the various embodiments .

With reference to Fig . 1 , the following will describe a system S for monitoring an overhead contact line OLE ; such system S comprises the following elements : - an apparatus 1 for monitoring said overhead contact line OLE ( e . g . an embedded device , an industrial PC, a development board, or the like ) , preferably positioned aboard an electric railway locomotive T which comprises a pantograph P that can collect an electric current from said overhead contact line ;

- video acquisition means 2 ( e . g . a digital optical sensor associated with a lens ) , preferably positioned within a container with IP67 protection, which preferably comprises also a visible and/or infrared light illuminator having a 850 and/or 940 nanometre wavelength, wherein said video acquisition means 2 are in communication with said apparatus 1 and are positioned in such a way as to frame a possible point of contact between said overhead contact line OLE and said pantograph P when the electric railway locomotive T is in an operating condition;

- a supervision computer 3 ( e . g . a remote physical server and/or a remote and/or " in-cloud" virtuali zed server ) in communication with said apparatus 1 and configured for receiving a signal SM generated by said apparatus 1 , and for noti fying an operator (preferably in real time ) when a monitoring datum is encoded in said signal SM which indicates that the apparatus 1 has detected an abnormal operating condition of the overhead contact line OLE and/or of the pantograph P .

An operating condition is considered herein to be abnormal when, although a certain point of the overhead contact line OLE and the pantograph P can still supply power to the electric railway locomotive T ( or a tram, a trolley bus , or the like ) , the continuation of such operating condition might lead to failure of said point of the overhead contact line OLE , e . g . said overhead contact line OLE might break and/or fall to the ground after a small number of runs ( 5- 10 ) , thus preventing circulation on at least a portion of the transportation network . It must be highlighted that the system S reduces the risks to which passengers are exposed; in particular, both the passengers waiting on the platform and those aboard the train might be hit by falling portions of the overhead line and hence undergo electric shock .

Also with reference to Fig . 2 , the apparatus 1 preferably comprises the following components :

- processing and/or control means 11 , e . g . one or more CPUs , GPUs , DSPs , FPGAs and/or the like , which preferably implement , whether in hardware and/or software form, a neural network trained by inputting to said neural network at least one training image representing a second overhead contact line and/or a second pantograph in an abnormal operating condition, and by forcing it to output a second monitoring datum indicating the occurrence of an abnormal operating condition;

- volatile memory means 12 , e . g . a random access memory RAM, in signal communication with the processing and/or control means 11 . When the apparatus 1 is in an operating condition, said volatile memory means 12 store the images acquired by the video acquisition means 2 ;

- non-volatile memory means 13 , preferably one or more magnetic disks (hard disks ) or a Flash memory or another type of memory, in signal communication with the processing and/or control means 11 and with the volatile memory means 12 , and wherein said non-volatile memory means 13 contain at least one set of instructions implementing a method for monitoring an overhead contact line OLE of a transportation network according to the invention . Moreover, said non-volatile memory means 13 may also contain information that makes it possible to configure and/or operate the neural network ( e . g . a set of internal weights , numbers of levels in the various layers , or the like ) ; - input and/or output (I/O) means 14 which can be connected to the video acquisition means 2 and which are configured for acquiring at least one image, which represents a possible point of contact between said overhead contact line OLE and the pantograph P that can collect an electric current from said overhead contact line. These input/output means 24 may comprise, for example, a USB™, IEEE 1394, RS232, RS485, IEEE 1284 adapter or the like;

- transmission means 15, preferably wired and/or wireless ones (e.g. a network interface such as 803.2 (also known as Ethernet) or 802.11 (also known as WiFi™) or 802.16 (also known as WiMax™) or a data network interface such as GSM/GPRS/UMTS/LTE/5G, TETRA™, LoRa™, XBee™, ZigBee™ or the like) , configured for transmitting a signal SM, in which said at least one monitoring datum is encoded, to the supervision computer 3, wherein said monitoring datum represents a state of (at least a portion of) the overhead contact line OLE;

- a communication bus 17 allowing information to be exchanged among the processing and/or control means 11, the volatile memory means 12, the non-volatile memory means 13, the input/output means 14, and the transmission means 15.

Also with reference to Fig. 3, when the apparatus 1 is in an operating condition, said apparatus 1 executes a set of instructions implementing the method according to the invention, which comprises the following phases:

- an acquisition phase Pl, wherein at least one image is acquired, via the input means 14, which represents the point of contact between said overhead contact line OLE and the pantograph P that can collect an electric current from the overhead contact line OLE;

- a processing phase P2, wherein at least one monitoring datum is determined (preferably by executing a process scheduled to comply with time constraints, i.e. a so-called "real-time" process) by means of the neural network and on the basis of said at least one image, wherein said monitoring datum represents the state of said overhead contact line OLE,

- a transmission phase P3, wherein the signal SM, in which said at least one monitoring datum is encoded, is transmitted, via transmission means 15, to the supervision computer 3.

This makes it possible to determine the occurrence of abnormal operating conditions as soon as a train comprising the apparatus 1 or any device implementing the method of the invention runs along the line, without requiring human intervention. It will thus be possible to immediately take the necessary steps and/or to carry out preventive maintenance in order to reduce the probability that faults or even accidents might occur.

The operating data collected during the data acquisition phase Pl may be transmitted to entities close to the apparatus 1 (i.e. close to the electric railway locomotive T) , e.g. to portable data processing and displaying systems used by service personnel (e.g. smartphones, tablets or laptops) or to data collection devices (e.g. microcontrollers comprising communication means, such as a LoRa™ communication interface or the like) positioned on travelling vehicles (e.g. train carriages circulating along the line) and/or flying vehicles (e.g. remotely controlled vehicles performing surveillance tasks over the line) . Furthermore, the operating data may be transmitted to remote entities (e.g. one or more servers) located in a data processing centre, i.e. to the supervision computer 3.

In more detail, the apparatus 1 is preferably configured for transmitting (during the transmission phase P3) , via the transmission means 15, said at least one operating datum to said supervision computer 3, wherein said supervision computer 3 is configured for executing the following steps:

- receiving, via communication means comprised in said supervision computer 3 (and compatible with the transmission means 15 of the apparatus 1) , said at least one image; - generating a predictive maintenance plan for said overhead contact line OLE on the basis of a historic model (e.g. generated by means of laboratory stress tests, so that said historic model can numerically model the ageing of the parts making up the overhead contact line OLE) and said at least one image, e.g. by calculating a probability of failure of the overhead contact line OLE on the basis of an image representing a possible point of contact between said overhead contact line OLE and a pantograph P (which can collect an electric current from said overhead contact line OLE) and using, as a model, a neural network trained with a historic sequence of images of a second overhead contact line (similar to the overhead contact line OLE) acquired before said second overhead contact line fails.

It must be pointed out that "predictive maintenance plan" refers to the definitions contained in the UNI 13306 2018, UNI 10147 2003 and UNI 9910 specifications, i.e. a maintenance plan developed by predicting at least one instant of time when the life cycle of a component will end (e.g. when it will fail) on the basis of a historic model and data concerning said component and/or a similar one.

It will thus be possible to carry out predictive maintenance in order to reduce the probability that faults or even accidents might occur.

The neural network is preferably of the feed-forward type, e.g. a Convolutional Neural Network (CNN) , a Fully Connected Deep Neural Network (FC DNN) , or a Hierarchical Temporal Memory (HTM) , such as, for example, the classification algorithm known as "YOLO" (You Only Look Once) , i.e. a neural network having a number of inputs equalling the quantity of pixels of the image that has been acquired by the input means 14 and that must undergo the categorization process; said neural network must be trained beforehand by means of a training process, which is preferably carried out by using a workstation comprising a CPU having more computational capacity than the processing and/or control means 11 , wherein said CPU is preferably configured for executing a set of instructions implementing a training algorithm, preferably a Stochastic Gradient Descent ( SGD) algorithm .

This makes it possible to determine , in real time , the occurrence of abnormal operating conditions the very first time a train runs along the line . It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

Also with reference to Fig . 4 , the following will describe some images which are used in the process of training said neural network and which are acquired by a video sensor comprised in the video acquisition means 2 or other video acquisition devices .

It must be pointed out that Figure 4 contains nine images , each one representing a particular abnormal operating condition . However, this set of abnormal situations should be considered as merely illustrative and non-limiting . Also , it must be pointed out that each image has been marked with a letter of the alphabet and contains an arrow and a few lines that have been added to highlight that portion of the photograph where the abnormal condition is shown; therefore , neither the letters nor the added graphics belong to the images used for training the neural network .

Figure 4 contains the following images :

- image A, which shows a first abnormal condition in which the second pantograph stands too low;

- image B, which shows a second abnormal condition in which the second pantograph stands too high;

- image C, which shows a third abnormal condition in which an electric arc is formed at a possible point of contact between the second pantograph and a second overhead contact line ;

- image D, which shows a fourth abnormal condition in which a dewirement of the second overhead contact line has occurred, so that the second pantograph cannot draw current from said overhead contact line ;

- image E , which shows a fi fth abnormal condition in which there is also a third ( additional ) overhead contact line and the second pantograph is only in contact with one overhead contact line ;

- image F, which shows a sixth abnormal condition in which a lubricant on the second pantograph and/or on said second overhead contact line has caught fire ;

- image G, which shows a seventh abnormal condition in which a contact crossbar that should mutually connect the two overhead contact lines is broken;

- image H, which shows an eighth abnormal condition in which a portion of the second pantograph in contact with the second overhead contact line ( also known as " sliding contact" ) has an abnormal orientation because said second pantograph has suf fered damage ;

- image I , which shows a ninth abnormal condition in which a contact force between the second pantograph and the second overhead contact line has an intensity below or above a range which is considered as normal .

In more detail , the method according to the invention preferably comprises also a training phase , wherein, prior to executing the processing phase P2 , said neural network is trained in a manner such that , when inputting to said neural network said at least one training image , the latter will output said second monitoring datum . This will allow the monitoring system S to be flexibly adapted in the course of its service li fe , e . g . by increasing the number of abnormal operating conditions that said monitoring system S will be able to autonomously recognise .

It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

As already partially described above , during the training phase , said at least one training image represents the second pantograph standing too low compared with a normal operating condition, and said second monitoring datum indicates that said second pantograph is standing too low .

This makes it possible to detect an abnormal operating condition ( and to transmit information describing such situation to a control ground station) due to insuf ficient tension of a catenary that supports the overhead contact line OLE , so that said overhead contact line OLE and/or another portion of the transportation network can be subj ected to preventive maintenance before a fault and/or an accident occurs . It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

In combination with or as an alternative to the above , during the training phase , said at least one training image represents the second pantograph standing too high compared with a normal operating condition, and said second monitoring datum indicates that said second pantograph is standing too high .

This makes it possible to detect an abnormal operating condition due to excessive tension of the catenary that supports the overhead contact line OLE and/or to sinking of the rails or the road ( e . g . because of a collapse ) , so that said overhead contact line OLE can be subj ected to preventive maintenance before a fault and/or an accident occurs . It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

In combination with or as an alternative to the above , during the training phase , said at least one training image represents an electric arc in proximity to a possible point of contact between the second pantograph and the second overhead contact line , and wherein said second monitoring datum indicates that an electric arc has formed between the second pantograph and the second overhead contact line .

This makes it possible to detect an abnormal operating condition due to ( temporarily) imperfect contact between the pantograph P and the overhead contact line OLE caused by incorrect tensioning and/or holding in position of said overhead contact line OLE and/or by a problem suf fered by the pantograph P . It must be pointed out that this situation is abnormal because the formation of an electric arc generates plasma that , having a temperature of thousands of Kelvin degrees , may damage the surfaces of the overhead contact line OLE and of the pantograph P, and may even break the contact line OLE by fusion and/or evaporation and/or sublimation; moreover, plasma also generates radio- frequency emissions in a very broad frequency spectrum, thus ( temporarily) preventing the radio devices aboard the electric railway locomotive T from communicating . It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

In combination with or as an alternative to the above , during the training phase , said at least one training image represents the second overhead contact line separated from the second pantograph, and wherein said second monitoring datum indicates that a dewirement has occurred .

This makes it possible to detect an abnormal operating condition due to a problem suf fered by one of the components making up the overhead contact line OLE , e . g . a contact wire , a carrying/ supporting cable , one or more droppers , a continuity j umper, a tensioning device , one or more insulators , one or more masts , a ground conductor, a (position or tension) registration element , or the like .

It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

In combination with or as an alternative to the above , during the training phase , said at least one training image represents also a third overhead contact line and the second pantograph is only in contact with one overhead contact line , and wherein said second monitoring datum indicates that a contact problem has occurred .

This makes it possible to detect an abnormal operating condition due to a contact problem that might produce electric arcs and damage both the overhead contact line OLE and the pantograph P because of the high temperatures generated . It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

In combination with or as an alternative to the above , during the training phase , said at least one training image represents a flame in proximity to the pantograph P and/or said second overhead line , and wherein said second monitoring datum indicates that a lubricant on the pantograph P and/or said second overhead line has caught fire .

This makes it possible to detect an abnormal operating condition which is also due to a contact problem producing small electric arcs that can ignite a lubricant spread on the overhead contact line OLE for the purpose of reducing friction with the pantograph P . Such flames may damage said overhead contact line OLE and/or said pantograph, leading to the formation of a layer of carbonised material on the overhead contact line OLE and/or on said pantograph, thus impairing the performance thereof in terms of electric conductivity . It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

With reference to Fig . 5 , and in combination with or as an alternative to the above , during the training phase , said at least one training image represents a broken contact crossbar CCB that should mutually connect two overhead contact lines ( a main overhead contact line OLE1 and a transverse overhead contact line OLE2 ) , and wherein said second monitoring datum indicates that said contact crossbar is broken .

This makes it possible to detect an abnormal operating condition that might cause abnormal movements of the overhead contact line OLE and/or of an additional overhead contact line because said overhead contact lines are no longer mutually connected . It will thus be possible to service the overhead contact lines sooner in order to reduce the probability that faults or even accidents might occur .

In combination with or as an alternative to the above , during the training phase , said at least one training image represents a portion of the second pantograph, which is in contact with the second overhead contact line , oriented di f ferently than in a normal operating condition, and wherein said second monitoring datum indicates that said portion of the second pantograph has an abnormal orientation .

This makes it possible to detect an abnormal operating condition in which the pantograph P has suf fered damage and may in turn damage the overhead contact line OLE , e . g . by tearing it of f and/or causing it to fall to the ground . It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

In combination with or as an alternative to the above , during the acquisition phase Pl , a force datum is also acquired, via the input means 14 , which represents at least an intensity of a force , preferably a force of contact with the overhead contact line ( composed of a static force , an aerodynamic force and an inertial force ) , acting upon the pantograph P, and wherein, during the processing phase P2 , said at least one monitoring datum is determined, by means of said neural network, also on the basis of said force datum . This makes it possible to constantly monitor the forces to which the pantograph P is subj ected, so as to be able to identi fy any anomalies in the contact between said pantograph P and said overhead contact line OLE , which might cause overheating, electric arcs or the like , which in turn might result in permanent damage to said overhead contact line OLE .

It will thus be possible to service the overhead contact line OLE sooner in order to reduce the probability that faults or even accidents might occur .

In combination with the above , during the training phase , said neural network is trained by inputting to said neural network also a second force datum representing an intensity of a force acting upon the second pantograph, wherein said intensity is higher or lower than a range considered as normal , and wherein said second monitoring datum indicates that said contact force has an abnormal intensity .

This makes it possible to adapt the monitoring system S at best to a speci fic overhead contact line , so as to be able to constantly monitor the forces applied to said overhead contact line OLE by the pantograph P, automatically detecting any abnormal operating conditions in regard to the forces acting upon the above-mentioned elements . It will thus be possible to service the overhead contact line OLE sooner should an anomaly of this kind occur, so as to reduce the risks of faults or even accidents .

In addition to the above , the processing and/or control means may also be configured for executing numerical processing techniques in order to correct the force measurements , e . g . by determining the active forces also as a function of the route conditions ( e . g . straight section and/or curve ) and of the vehicle ' s travel conditions ( e . g . inclination and/or speed and/or acceleration of the vehicle travelling along the line ) . In other words , the input means 14 are preferably also configured for acquiring ( during the acquisition phase Pl ) at least one travel datum which represents ( in numerical format ) at least an inclination and/or a speed and/or an acceleration to which said apparatus 1 is subj ected, and wherein the processing means 11 are also configured for executing ( during the processing phase P2 ) the following steps :

- determining, on the basis of said at least one image ( acquired during the acquisition phase Pl ) , a route condition datum indicating i f said possible point of contact between said overhead contact line OLE and said pantograph P is over a curve of said transportation line , e . g . by using a second neural network similar to the previously described neural network and trained to recognise i f a portion of a second overhead line suspended along the route is over a curve or a straight section of a transportation network;

- determining a third force datum on the basis of said second force datum and said travel datum, when said route condition datum indicates that said overhead contact line OLE and said pantograph P are over a curve of said transportation network, e . g . by multiplying said second force datum by a correction factor determined by means of a trigonometric calculation quanti fying the force portion due to the dynamic loads to which the pantograph P is subj ected;

- determining said at least one monitoring datum on the basis of said third force datum .

This improves the precision of the monitoring system S , resulting in a more accurate ( and automatic ) monitoring of the forces applied to said overhead contact line OLE by the pantograph P . It will thus be possible to service the overhead contact line OLE sooner should an anomaly occur, so as to reduce the probability that faults or even accidents might occur .

In addition to the above , the input means are preferably configured for acquiring at least one position datum which represents a position of said apparatus 1 ( e . g . acquiring a position datum from a GNSS receiver comprised in said apparatus 1 and/or in the electric railway locomotive T and/or by odometrical detection) , and wherein the processing means 11 are preferably also configured for executing (during the processing phase) the following steps:

- determining a line condition datum on the basis of said at least one position datum (e.g. using a map associating a particular line condition datum with at least one area) , wherein said line condition datum represents a wear condition of said overhead contact line OLE, e.g. as a fixed-point or floating-point numerical value ranging from 1.00 (new overhead line) to 0.10 (worn-out overhead line) ;

- determining, by means of said neural network, said at least one monitoring datum also on the basis of said line condition datum, e.g. by multiplying the force datum by the line condition datum.

This further improves the adaptation of the monitoring system S to a specific overhead contact line, since the wear of the overhead contact line OLE is also taken into account, considering also the interaction between the pantograph and the overhead contact line OLE, which will result in a further reduction of the probability that faults or even accidents might occur.

Of course, the example described so far may be subject to many variations .

Some of the possible variants of the invention have been described above, but it will be clear to those skilled in the art that other embodiments may also be implemented in practice, wherein several elements may be replaced with other technically equivalent elements. The present invention is not, therefore, limited to the above-described illustrative examples, but may be subject to various modifications, improvements, or replacements of equivalent parts and elements without however departing from the basic inventive idea, as specified in the following claims.