INVENTION FIELD

The present invention relates to a method for identifying the operating state of an industrial machinery, in particular of hydraulic machines and systems.

In particular, the present invention relates to predictive maintenance applied to hydraulic machines and systems.

KNOWN PRIOR ART

As is known, predictive maintenance is a type of preventive maintenance that is carried out following the identification of one or more parameters that are measured and processed using appropriate mathematical models in order to identify the remaining time before the failure.

During the last few years strongly encouraged by Industry 4.0, an intense activity has developed aimed at collecting data and information from components and machinery.

Specifically, measurement systems (sensors) have been created that report the information/data within a viewer capable of making them available.

In detail we talk about the measurement of multiple quantities and/or phenomena: temperatures; levels; flows; vibrations, distances, etc. All made for general applications in an unorganized way.

The obvious disadvantages are, firstly, that the data are expressed as a cold number not always easy to interpret. A further disadvantage is that the data are not analysed and interpreted for individual and specific applications.

The prior art also requires project infrastructures to be created for each individual application.

Not least among the disadvantages of the prior art are the fact that the data are made available in the most diverse forms and therefore not coordinated with each other, the fact that the data remain ends in themselves and the fact that in fact the flow of the data stops at the simple display and consultation.

As for the current work organization, even in the presence of scheduled maintenance procedures, today we often work in emergency with very high costs and it is often not possible to optimize the purchase of spare parts.

In addition, scheduled maintenance requires the replacement of parts and components even if not completely worn.

An object of the present invention is to ensure that all the information collected by components and machinery is the basis for being able to identify, plan and carry out maintenance activities.

Another object of the invention is to overcome the logic and limitation of scheduled maintenance procedures.

A further object of the invention is to provide a solution that is simple to manufacture and with low manufacturing costs.

SUMMARY OF THE INVENTION

The present invention aims to achieve the purposes described above by a method for identifying the operating state of an industrial machinery, in particular of hydraulic machines and systems, wherein said method comprises at least the following steps:

- use of at least one sensor to collect data relating to the operating status of a hydraulic machine or system;

- prediction over time of the state of wear of components belonging to hydraulic machines or systems; wherein - said phase of prediction over time of the state of wear of components belonging to machines or hydraulic systems is carried out by analyzing the data detected by the at least one sensor using Machine Learning techniques.

The present invention is in fact capable of collecting data from the field in an independent, organized and optimized manner - sending it to the cloud or to a specific local section - viewing it in real time with specific indicators - contextualizing and analyzing it by application.

The machine-learning activity assisted by artificial intelligence and (specific) knowhow provides predictions over time on the state of wear of components and machinery.

This prediction is used as a basis for identifying the maintenance activities to be performed, their priority and relative planning. The data collected from the field is sent (independently) to an artificial intelligence located in the cloud or locally that processes it.

Thus, the advantages of the present invention are:

The possibility of implementing the monitoring system in a simple and non-invasive way, applicable everywhere, for any application, where instead currently the infrastructures are to be created for each specific case.

The invention also makes it possible to have targeted information available on your system/application/sector in real time.

Currently the information is generic - not contextualised and analysed by single application or technical specialisation.

The invention also allows the analysis of data available for specific application, where currently the information is not analyzed or processed and compared.

The invention also offers the possibility of predicting the state of wear and functional decay of components and machines over time. In fact, preventing failure, as well as the possibility of planning (even at pre-planned time intervals e.g.: 90/180 days, etc.) correctly maintenance activities by optimizing the operation itself, the flow of management of spare parts and increasing safety.

The invention offers further possibility of increasing the productivity of the machines, today penalised by sudden failures.

The invention also makes it possible to reduce the general costs attributable to maintenance and to be able to operate in an organized and planned manner, allowing drastically reducing costs.

Moreover, thanks to the invention, the components are replaced only when strictly necessary.

Last but not least, it should be noted that the invention makes it possible to prevent sudden failures, which means increasing safety and environmental sustainability, where - as is well known - sudden failures can compromise the health of operators, for example through spills of fluids in the event of breakages, which produce pollution.

There are different versions of the hardware part, different solutions for data management and prediction.

Another absolute innovation, in addition to the simplicity of installation and flexibility of the hardware part, is represented by the fact that the invention is modular and expandable up to the coverage of the entire machine.

In one variant, the hardware for implementing the method may also be made in a “mobile” version transportable and/or positionable as needed.

Further characteristics of the invention can be deduced from the dependent claims.

BRIEF DESCRIPTION OF THE FIGURES Further characteristics and advantages of the invention will become more evident in the light of the detailed description which follows with the aid of the accompanying drawing tables in which:

- figure 1 illustrates a device for identifying the operating state of hydraulic machines and systems, according to an embodiment of the present invention; and

- figure 2 illustrates a display belonging to the device of figure 1 .

DETAILED DESCRIPTION OF SOME EMBODIMENTS OF THIS INVENTION

The invention will now be described with initial reference to the figures where in particular, in figure 1 a device for identifying the operating state of hydraulic machines and systems is visible, according to an embodiment of the present invention and globally indicated with the numerical reference 10.

The device 10, which comprises a frame 15, possibly mounted on wheels 17 for greater mobility, also comprises a display 20 illustrated in more detail in figure 2.

More specifically, the display 20 is configured to report a plurality of information relating to power oil and lubrication present in the various machines of the monitored system.

In particular, there is a display 30 that indicates the oil temperature in real time, for example expressed in degrees Celsius or Fahrenheit, depending on the applications

A display 40 is also available indicating the degree of oil contamination.

There is also a display that indicates a parameter related to the quality of the oil, in particular the HPN parameter.

Each of the above parameters can be measured by means of appropriate sensors installed on machines and systems on which it is desired to monitor the hydraulic state.

Among the parameters of interest by way of example are: Chemical/physical parameters - as in the case of oil, dimensions of parts of the machine to be monitored, vibrations of the same, sounds/ultrasounds emitted, visual survey instruments, temperatures, thermocameras for infrared study of the machine and/or x-rays, it is always possible to add parameters/sensors/evaluation means to improve the evaluation diagnostics related to this method, obviously remaining within the objects of the present invention.

The processing of the data thus collected allows to perform scheduled maintenance events.

There are also several buttons indicated with the numerical references 60 for the data button, 70 for the alarm button and 80 for the data analysis button.

The invention makes it possible to implement a method for identifying the operating state of an industrial machinery, in particular of hydraulic machines and systems, wherein the aforementioned method comprises at least the following steps:

- use of at least one sensor to collect data related to the operating status of a hydraulic machine or system;

- prediction over time of the state of wear of components belonging to hydraulic machines or systems; wherein

- the aforementioned phase of prediction over time of the state of wear of components belonging to machines or hydraulic systems is carried out by analyzing the data detected by the at least one sensor using Machine Learning techniques.

For the purposes of this description, it should be noted that the term Machine Learning refers to a branch of artificial intelligence that collects a set of methods, such as computational statistics, pattern recognition, artificial neural networks, filtering, convolutional neural networks, to progressively improve performance and flexibility of intervention in the industrial field.

In more discursive terms, Machine Learning is the science that allows computers to learn and act as humans do, and improve their learning over time autonomously, providing them with data and information in the form of observations and interactions with the real world.

Such techniques may include:

Artificial Neural Networks (ANN) Convolutional Neural Networks - CNN

Hierarchical Temporal Memory and several others

There are several approaches to machine learning including:

Supervised learning: Inputs and outputs provided by the operator

Unsupervised learning: Linked to the autonomous recognition of patterns in data

Reinforcement learning: learning technique in which the machine learns from the consequences of its actions. Essentially the machine learns through trial and error.

In the method of the invention, the phase of predicting over time the state of wear of components belonging to hydraulic machines or systems is also used as a basis for identifying the maintenance activities to be performed, their priority and their relative planning.

A further absolute innovation, in addition to the simplicity of installation and flexibility of the hardware part, is represented by the fact that the present invention operates as a modular and expandable system up to the cover of the entire machine, therefore the prediction activity is carried out on the oil as an important element of the hydraulic system but also on all the other components.

The present invention is further articulated on at least three different phases at 20 increasing complexity, without implying that the number of such phases is limited and excludes other possible implementations as defined in the claims appended hereto in description.

First, the aforementioned phase of predicting over time the state of wear of components belonging to hydraulic machines or systems is essentially based on the analysis of the hydraulic fluid.

This phase can also be synthetically defined as ENTRY, i.e. predictive management based on hydraulic fluid analysis. Alternatively, the phase of predicting over time the state of wear of components belonging to hydraulic machines or systems is based on the analysis of all the hydraulic components present in the system, including oil.

This phase can also be defined synthetically as ILLIMITY, i.e. predictive management based on the analysis of all the hydraulic components present in the system, including oil.

It is also possible to use a briefly defined as OPEN TO, i.e., in addition to the predictive management of the hydraulic system, it offers the possibility of involving other disciplines such as mechanics, lubrication, etc.

Note that the phase of predicting over time the state of wear of components belonging to machines or hydraulic systems, is based on the analysis and visualization by use of a control panel on board of the machine that allows to display hydraulic parameters in real time, said hydraulic parameters being chosen among temperature, contamination, quality indexes (HPNumber) of the oil or combinations thereof.

Naturally, amendments or improvements may be made to the invention as described, dictated by contingent or particular reasons, without thereby departing from the scope of the invention.