A vehicle running gear on-board fault online monitoring and diagnosis system, device, equipment and storage medium

By dividing the area of ​​the wheels and axles, using machine learning algorithms to establish diagnostic models, monitor the vehicle operating status in real time, and identify fault characteristic points, the problems of slow fault diagnosis and insufficient prediction capabilities in the existing technology are solved, and the vehicle traveling department is realized Accurate prediction and early warning of faults improve maintenance efficiency and safety.

CN118849980BActive Publication Date: 2024-12-03HANGZHOU ZHONGGANG METRO EQUIP MAINTENANCE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411338748.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-25
Publication Date
2024-12-03
Estimated Expiration
2044-09-25

AI Technical Summary

Technical Problem

The existing fault monitoring technology of the vehicle travel department is slow to respond, making it difficult to achieve real-time or near-real-time fault diagnosis, lacks efficient data processing and analysis capabilities, and cannot identify complex fault patterns early, resulting in increased maintenance costs and driving safety hazards.

Method used

By dividing the wheels and axles, collecting historical fault and operating status data, using machine learning algorithms to establish diagnostic models, monitoring operating status in real time, identifying abnormal indicators and fault feature points, and building a neural network model for failure probability prediction.

Benefits of technology

Significantly improve the accuracy of fault detection, realize the prediction of future faults of the vehicle traveling department, reduce downtime and maintenance costs, provide preventive maintenance, and improve vehicle operation efficiency and safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118849980B_ABST
    Figure CN118849980B_ABST
Patent Text Reader

Abstract

The present invention provides a vehicle running gear on-board fault online monitoring and diagnosis system, device, equipment and storage medium, the system performs: divide each wheel to form wheel sub-areas; divide the axle into regions to form axle sub-areas; obtain the historical fault data of each wheel and axle sub-area; obtain the historical operating status data of each wheel and axle sub-area respectively; obtain the ratio of historical fault data to operating status data based on the same wheel and axle sub-area, and define the ratio as the historical fault probability; learn and train the historical fault probability of each wheel sub-area and each axle sub-area respectively to obtain a diagnosis model; diagnose the future fault probability respectively through the diagnosis model based on a preset number of diagnosis steps. The present invention can monitor and diagnose faults in real time, effectively improve the accuracy and efficiency of fault detection, reduce maintenance costs and improve driving safety.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle-mounted fault monitoring and diagnosis, and more specifically, to a vehicle running gear on-board fault online monitoring and diagnosis system, device, medium and computing equipment. Background Art

[0002] Existing vehicle running gear fault monitoring technologies mainly rely on traditional sensors and manual inspection methods, which can often only identify faults after they occur and lack the ability to perform preventive maintenance. The technical principles are usually based on real-time monitoring of the vehicle's operating status, collecting data through sensors, and then identifying anomalies through threshold judgment or simple algorithm analysis. However, these methods have obvious limitations in fault prediction, mainly manifested in the lack of in-depth analysis and application capabilities of data, making it difficult to achieve early identification and early warning of complex fault modes.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: first, the traditional fault monitoring system is often slow to respond and cannot achieve real-time or near real-time fault diagnosis; second, due to the lack of efficient data processing and analysis capabilities, it is difficult for the existing system to accurately identify fault signs from a large amount of operating data; finally, the prior art has limited ability in fault prediction and can often only respond passively after a fault occurs, lacking the ability to actively prevent and maintain. These problems lead to increased maintenance costs and hidden dangers to driving safety. Summary of the invention

[0004] The invention provides a vehicle running gear on-board fault online monitoring and diagnosis system, device, medium and computing equipment.

[0005] In a first aspect of the present invention, a vehicle running gear on-board fault online monitoring and diagnosis system is provided, comprising:

[0006] Dividing each wheel into regions respectively, so that each wheel forms at least two wheel sub-regions;

[0007] Dividing the axle into regions to form at least two axle sub-regions;

[0008] Acquire a number of historical fault data of each wheel sub-area and each axle sub-area based on a preset time interval;

[0009] Based on the preset time interval, respectively obtain a number of historical operating status data of each wheel sub-region and each axle sub-region;

[0010] Acquire a ratio of the historical fault data to the historical operating state data based on the same wheel sub-region or the same axle sub-region, and define the ratio as a historical fault probability;

[0011] The historical failure probability of each wheel sub-area and each axle sub-area is learned and trained by a machine learning algorithm to obtain a diagnosis model;

[0012] The future failure probability of each wheel sub-region and each axle sub-region is diagnosed respectively through the diagnosis model based on a preset number of diagnosis steps.

[0013] Furthermore, a plurality of historical fault data of each wheel sub-region and each axle sub-region are obtained based on a preset time interval, including:

[0014] Monitor the running status of the vehicle running gear in real time through sensors;

[0015] Obtaining abnormal indicators of the operating status through a data analysis algorithm;

[0016] Determine whether each wheel sub-region or each axle sub-region has the abnormal indicator;

[0017] If yes, then record the fault data once for each wheel sub-area or axle sub-area having the abnormal indicator;

[0018] All historical fault data of each wheel sub-region and each axle sub-region are acquired respectively based on the preset time interval.

[0019] Furthermore, the abnormal indicators of the operating state are obtained by a data analysis algorithm, including:

[0020] Acquiring operating parameters of the vehicle running gear by sampling at a first preset frequency;

[0021] Predicting a preset number of fault characteristic values ​​based on each sampling parameter;

[0022] Defining the fault characteristic value to have the highest confidence level;

[0023] Obtain the fault feature value with the highest confidence level among all confidence levels as the key fault feature;

[0024] Calculating the correlation between the key fault characteristic value and the fault characteristic value;

[0025] The key fault characteristic values ​​with a correlation greater than or equal to a preset ratio are retained as abnormal indicators;

[0026] Gets a collection of all anomaly indicators.

[0027] Furthermore, based on the preset time interval, a plurality of historical operating status data of each wheel sub-region and each axle sub-region are respectively obtained, including:

[0028] Acquiring the running status data of the vehicle running gear through continuous monitoring;

[0029] Acquire fault feature points in the operating status data through a status recognition algorithm;

[0030] Determine whether each wheel sub-region or each axle sub-region has the fault characteristic point;

[0031] If so, the wheel sub-region or the axle sub-region where the fault characteristic point exists respectively records the running status data once;

[0032] All historical operating status data of each wheel sub-area and each axle sub-area are respectively acquired based on the preset time interval.

[0033] Furthermore, obtaining the fault characteristic points in the operating status data through a status recognition algorithm includes:

[0034] Standardizing the operating status data to obtain standardized status data;

[0035] Use key operating parameter analysis technology to identify potential fault feature points;

[0036] Determine the persistence and stability of fault feature points through time series analysis method;

[0037] Integrate the confirmed fault feature points to form a fault feature sequence.

[0038] Furthermore, the historical failure probabilities of each wheel sub-area and each axle sub-area are learned and trained by a machine learning algorithm to obtain a diagnostic model, including:

[0039] Generate a number of acquisition time points based on a preset time interval;

[0040] Based on all acquisition time points, the historical failure probability of each wheel sub-area and each axle sub-area is obtained to generate a data set;

[0041] All datasets were integrated to form a total dataset;

[0042] performing standardization on the total data set;

[0043] Dividing the standardized data set into a training set and a validation set according to a preset ratio;

[0044] Define the topology of the neural network model;

[0045] The training set is input into the neural network model for iterative training to obtain the diagnostic model corresponding to the minimum root mean square error.

[0046] Furthermore, the neural network model characterizes the specific calculation process of the failure probability through the following formula:

[0047] in, represents the failure probability of the output; Indicates The weight of the input features to the hidden layer; Indicates Input feature values; represents the bias of the hidden layer; Represents the activation function.

[0048] In a second aspect of the present invention, a vehicle running gear on-board fault online monitoring and diagnosis device is provided, comprising:

[0049] A wheel region division module, used for dividing each wheel into regions, each wheel forming at least two wheel sub-regions;

[0050] An axle region division module, used for dividing the axle into regions to form at least two axle sub-regions;

[0051] A historical fault data acquisition module, used to acquire a number of historical fault data of each wheel sub-region and each axle sub-region based on a preset time interval;

[0052] A historical operation status data acquisition module, used for acquiring a plurality of historical operation status data of each wheel sub-region and each axle sub-region based on the preset time interval;

[0053] A historical fault probability definition module, used for obtaining a ratio of the historical fault data to the historical operating status data based on the same wheel sub-region or the same axle sub-region, and defining the ratio as a historical fault probability;

[0054] A diagnostic model acquisition module, used to learn and train the historical failure probability of each wheel sub-region and each axle sub-region through a machine learning algorithm to obtain a diagnostic model;

[0055] The future fault probability diagnosis module is used to diagnose the future fault probability of each wheel sub-region and each axle sub-region respectively through the diagnosis model based on a preset number of diagnosis steps.

[0056] In a third aspect of the present invention, an electronic device is provided, comprising: at least one processor, a memory and an input-output unit; wherein the memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute any one of the methods described in the first aspect.

[0057] In a fourth aspect of the present invention, a computer-readable storage medium is provided, which includes instructions, and when the instructions are executed on a computer, the computer executes any one of the methods in the first aspect.

[0058] The above-mentioned embodiments of the present invention have at least the following beneficial effects: the vehicle running gear on-board fault online monitoring and diagnosis system of the present invention can significantly improve the accuracy of fault detection through fine regional division and in-depth analysis of historical data. The system can establish a diagnostic model based on historical fault probabilities and operating status data using machine learning algorithms, thereby predicting the future fault probability of the vehicle running gear. This predictive capability makes maintenance work more targeted and timely, reducing downtime and maintenance costs caused by sudden faults.

[0059] In addition, the system can detect and warn potential faults in a timely manner through real-time monitoring and intelligent analysis, thus providing strong support for preventive maintenance of vehicles. This proactive fault management method can not only improve the operating efficiency and reliability of the vehicle, but also enhance driving safety and provide drivers and passengers with a more secure travel experience. Through continuous learning and optimization, the system can adapt to different operating environments and conditions, ensuring accurate fault diagnosis results in all situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0061] Figure 1 A schematic diagram of a flow chart of an on-board fault online monitoring and diagnosis system for a vehicle running gear provided by an embodiment of the present invention;

[0062] Figure 2 A schematic diagram of the structure of an on-board fault online monitoring and diagnosis device for a vehicle running gear provided by an embodiment of the present invention;

[0063] Figure 3 The schematic diagram schematically shows the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0065] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, apparatus, method or computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0066] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0067] Reference below Figure 1 , Figure 1 This is a flow chart of a vehicle running gear on-board fault online monitoring and diagnosis system provided by an embodiment of the present invention. Figure 1 As shown, a vehicle running gear on-board fault online monitoring and diagnosis system 100 includes:

[0068] Step 101, dividing each wheel into regions, so that each wheel forms at least two wheel sub-regions;

[0069] Step 102, dividing the axle into regions to form at least two axle sub-regions;

[0070] Step 103, obtaining a number of historical fault data of each wheel sub-region and each axle sub-region based on a preset time interval;

[0071] Step 104, acquiring a plurality of historical operating status data of each wheel sub-region and each axle sub-region based on the preset time interval;

[0072] Step 105, obtaining a ratio of the historical fault data to the historical operating status data based on the same wheel sub-region or the same axle sub-region, and defining the ratio as a historical fault probability;

[0073] Step 106, learning and training the historical failure probability of each wheel sub-region and each axle sub-region by a machine learning algorithm to obtain a diagnosis model;

[0074] Step 107 , diagnose the future failure probability of each wheel sub-region and each axle sub-region respectively through the diagnosis model based on a preset number of diagnosis steps.

[0075] It should be noted that the wheel sub-areas and axle sub-areas in the system refer to several monitoring areas into which the wheels and axles are divided. These areas can be physically divided or logically distinguished. The historical fault data and historical operating status data of each area can be collected by sensors, such as temperature sensors, vibration sensors, etc. to monitor the real-time status of the vehicle running parts. The preset time interval can be fixed, such as collecting data once a minute or every hour, or it can be dynamically adjusted and optimized according to the vehicle's usage and historical fault data.

[0076] Preferably, the online fault monitoring and diagnosis system may further include the following steps: First, real-time data collection is performed on each sub-area of ​​the wheel and axle through high-precision sensors. These sensors may include but are not limited to pressure sensors, speed sensors, and acceleration sensors. Secondly, after data collection, data analysis algorithms are used to identify abnormal indicators of the collected data. These algorithms may be based on statistical principles or pattern recognition techniques. In addition, the system may set up an adaptive learning mechanism to continuously optimize the calculation model of the fault probability based on historical data to improve the accuracy of diagnosis. Finally, the system may provide a user interface that allows the operator to adjust the monitoring parameters and view the fault diagnosis results as needed.

[0077] In some embodiments, a plurality of historical fault data of each wheel sub-region and each axle sub-region are obtained based on a preset time interval, including:

[0078] Monitor the running status of the vehicle running gear in real time through sensors;

[0079] Obtaining abnormal indicators of the operating status through a data analysis algorithm;

[0080] Determine whether each wheel sub-region or each axle sub-region has the abnormal indicator;

[0081] If yes, then record the fault data once for each wheel sub-area or axle sub-area having the abnormal indicator;

[0082] All historical fault data of each wheel sub-region and each axle sub-region are acquired respectively based on the preset time interval.

[0083] It should be noted that real-time monitoring through sensors can refer to the use of a range of different types of sensors, such as temperature sensors, pressure sensors, vibration sensors, etc., to collect real-time operating data of the vehicle's running gear. Data analysis algorithms include statistical analysis, pattern recognition or machine learning techniques to identify abnormal patterns or indicators that may indicate a fault from the collected data. Abnormal indicators can be parameters that are beyond the normal operating range, such as excessively high temperatures, abnormal vibration frequencies or pressure changes.

[0084] Preferably, the online fault monitoring and diagnosis system can further refine the operation steps, for example: first, define the data collection frequency and threshold of each sensor to ensure the accuracy and timeliness of the data. Secondly, develop or select suitable data analysis algorithms that can adapt to different types of data and possible failure modes. In addition, an automated fault recording system can be set up. Once an abnormal indicator is detected, the system will automatically record the fault data and mark the relevant areas for further analysis and diagnosis. It is also possible to consider introducing an adaptive mechanism so that the system can continuously optimize the monitoring parameters and algorithms based on new data and diagnostic results, thereby improving the diagnostic accuracy and efficiency of the system.

[0085] In some embodiments, obtaining the abnormal indicator of the operating state through a data analysis algorithm includes:

[0086] Acquiring operating parameters of the vehicle running gear by sampling at a first preset frequency;

[0087] Predicting a preset number of fault characteristic values ​​based on each sampling parameter;

[0088] Defining the fault characteristic value to have the highest confidence level;

[0089] Obtain the fault feature value with the highest confidence level among all confidence levels as the key fault feature;

[0090] Calculating the correlation between the key fault characteristic value and the fault characteristic value;

[0091] The key fault characteristic values ​​with a correlation greater than or equal to a preset ratio are retained as abnormal indicators;

[0092] Gets a collection of all anomaly indicators.

[0093] It should be noted that the method of obtaining abnormal indicators of the operating status of the vehicle running gear through a data analysis algorithm involves sampling the operating parameters of the vehicle running gear, predicting fault characteristic values, defining key fault characteristics, and calculating the correlation between these characteristic values ​​to finally determine the abnormal indicators.

[0094] Specifically, the sampling of the first preset frequency refers to the system collecting data on the operating parameters of the vehicle running gear at a certain time interval, which can be every second, every minute or any suitable time interval. The fault feature value is based on the historical fault data and operating status data, and the algorithm predicts the parameter value that may indicate a fault. Confidence refers to the reliability index of these fault feature values, which is used to evaluate the accuracy of the predicted value. Key fault features refer to those with the highest confidence among all predicted fault feature values, which are considered to be key indicators of the occurrence of faults.

[0095] Preferably, the system can further refine the operation steps, including: first, determine the appropriate sampling frequency to ensure the representativeness and real-time nature of the data. Secondly, develop or select suitable algorithms to predict fault characteristic values, which may include time series analysis, statistical models, or machine learning models. Then, define a confidence assessment mechanism to quantify the reliability of each fault characteristic value. In addition, a threshold can be set to determine which fault characteristic values ​​have a high enough correlation to be considered as abnormal indicators. Finally, an integrated algorithm for abnormal indicators can be developed to integrate all key fault characteristic values ​​to form a comprehensive set of abnormal indicators for subsequent fault diagnosis.

[0096] In some embodiments, obtaining a plurality of historical operating status data of each wheel sub-region and each axle sub-region based on the preset time interval includes:

[0097] Acquiring the running status data of the vehicle running gear through continuous monitoring;

[0098] Acquire fault feature points in the operating status data through a status recognition algorithm;

[0099] Determine whether each wheel sub-region or each axle sub-region has the fault characteristic point;

[0100] If so, the wheel sub-region or the axle sub-region where the fault characteristic point exists respectively records the running status data once;

[0101] All historical operating status data of each wheel sub-area and each axle sub-area are respectively acquired based on the preset time interval.

[0102] It should be noted that the method for obtaining the historical operating status data of each wheel sub-region and axle sub-region of the vehicle running gear includes continuous monitoring, status identification and recording of fault feature points.

[0103] Specifically, continuous monitoring means that the system continuously collects real-time data from the vehicle's running gear, which may include parameters such as speed, temperature, vibration, etc. State recognition algorithms are computational methods used to analyze this data and identify characteristic points that may indicate a fault. Fault characteristic points are points found in the data that are abnormal or deviate from the normal operating mode, which may indicate potential faults.

[0104] Preferably, the system may further include the following operating steps, including:

[0105] First, determine the frequency and type of data collection to ensure that the operating status of the vehicle running gear can be fully captured. Secondly, develop or select suitable state recognition algorithms, which may be based on statistical analysis, pattern recognition or machine learning techniques to extract fault feature points from the data. Then, set the recognition criteria for fault feature points, for example, by setting thresholds to determine which data points are considered abnormal. In addition, an automated recording system can be introduced. Once the fault feature points are identified, the system will automatically record these data points and mark them for subsequent analysis and diagnosis. It is also possible to consider using multi-dimensional data fusion technology to combine data from different sensors to improve the accuracy and reliability of fault feature point identification.

[0106] In some embodiments, obtaining the fault feature points in the operating status data by using a status recognition algorithm includes:

[0107] Standardizing the operating status data to obtain standardized status data;

[0108] Use key operating parameter analysis technology to identify potential fault feature points;

[0109] Determine the persistence and stability of fault feature points through time series analysis method;

[0110] Integrate the confirmed fault feature points to form a fault feature sequence.

[0111] It should be noted that the specific method of obtaining fault characteristic points in the operating status data through the state recognition algorithm includes standardizing the operating status data, identifying potential fault characteristic points using key operating parameter analysis technology, determining the continuity and stability of the fault characteristic points through a time series analysis method, and integrating the confirmed fault characteristic points to form a fault characteristic sequence.

[0112] Specifically, the state recognition algorithm refers to a series of calculation methods used to analyze the operating state data and identify the fault mode. Standardization processing involves the normalization or standardization of data to eliminate the influence of different dimensions and magnitudes. Key operating parameter analysis techniques include feature extraction and pattern recognition, which are used to identify fault-related parameters from the data. Time series analysis methods are used to evaluate the change trend of fault feature points over time to determine whether they are persistent and stable.

[0113] Preferably, the system can further refine the operation steps, including: first, define the standardized processing flow of the data, for example, select an appropriate normalization method, such as Z-score standardization or Min-Max standardization. Secondly, develop or select suitable key operating parameter analysis techniques, which may be based on machine learning algorithms, such as support vector machines (SVM) or random forests, to identify fault feature points. Then, apply time series analysis methods, such as autoregressive integrated moving average (ARIMA) models, to evaluate the persistence and stability of fault feature points. In addition, an integrated algorithm can be developed to integrate all confirmed fault feature points to form a comprehensive fault feature sequence for subsequent fault diagnosis and prediction. It is also possible to consider introducing a feedback mechanism to continuously adjust and optimize the state recognition algorithm based on the diagnosis results to improve the accuracy and efficiency of fault identification.

[0114] In some embodiments, the historical failure probabilities of each wheel sub-region and each axle sub-region are learned and trained by a machine learning algorithm to obtain a diagnostic model, including:

[0115] Generate a number of acquisition time points based on a preset time interval;

[0116] Based on all acquisition time points, the historical failure probability of each wheel sub-area and each axle sub-area is obtained to generate a data set;

[0117] All datasets were integrated to form a total dataset;

[0118] performing standardization on the total data set;

[0119] Dividing the standardized data set into a training set and a validation set according to a preset ratio;

[0120] Define the topology of the neural network model;

[0121] The training set is input into the neural network model for iterative training to obtain the diagnostic model corresponding to the minimum root mean square error.

[0122] It should be noted that the preset time interval refers to the system collecting data within a fixed time period, such as every 10 minutes or every hour. The collection time point refers to the specific moment when the system actually collects data within these time intervals. The data set refers to the collection of all historical failure probability data collected at these time points. Normalization processing refers to converting data into a unified format or range for easy processing by machine learning models. The neural network model is a commonly used machine learning model that makes predictions by learning the relationship between input data and output results.

[0123] Preferably, the online fault monitoring and diagnosis system can further refine the operation steps, including: first, determine the appropriate preset time interval and collection time point to ensure the continuity and representativeness of the data. Secondly, select a suitable data normalization method, such as minimum-maximum normalization or Z-score normalization, to improve the training efficiency and accuracy of the model. Then, design the topology of the neural network model, including the number of layers, the number of neurons in each layer, and the selection of activation functions. In addition, techniques such as cross-validation can be used to evaluate the performance of the model, and strategies such as early stopping can be used to prevent overfitting. Finally, different optimization algorithms, such as gradient descent or its variants, can be explored to train the model to achieve the best diagnostic effect.

[0124] In some embodiments, the neural network model characterizes the specific calculation process of the failure probability through the following formula:

[0125] in, represents the failure probability of the output; Indicates The weight of the input features to the hidden layer; Indicates Input feature values; represents the bias of the hidden layer; Represents the activation function.

[0126] It should be noted that the topology of the neural network model refers to the configuration of each layer in the model, including the number of neurons in the input layer, hidden layer, and output layer. Minimum root mean square error refers to minimizing the difference between the predicted value and the actual value through iterative optimization algorithm during model training. The specific calculation process of the failure probability involves how the model converts input features into output failure probability.

[0127] Preferably, the following operation steps can be further refined: First, define the input features, which may include historical fault data, operating status data, etc., and determine the weight and bias of each feature. Secondly, select a suitable activation function, such as sigmoid or ReLU function, to increase the nonlinear expression ability of the model. Then, through methods such as cross-validation, determine the optimal model parameters and structure to improve the generalization ability of the model. In addition, regularization techniques, such as L1 or L2 regularization, can be used to prevent the model from overfitting.

[0128] More specifically, the following refinements or alternatives can be provided for the model training process: First, optimization algorithms such as batch gradient descent or stochastic gradient descent can be used to train the model to minimize the loss function. Second, a learning rate decay strategy can be set to ensure the stability and convergence of model training. In addition, the introduction of ensemble learning methods such as bagging or boosting can be considered to further improve the prediction accuracy of the model. Different neural network architectures, such as convolutional neural networks (CNNs) or recurrent neural networks (RNNs), can also be explored to adapt to different types of data features and failure modes.

[0129] The above-mentioned embodiments of the present invention have the following beneficial effects: by dividing the wheels and axles of the vehicle running gear into multiple sub-areas and collecting the historical fault and operating status data of these areas respectively, the system of the present invention can build a more detailed fault diagnosis model. This regionalized data collection method can improve the resolution of fault detection, allowing the system to conduct in-depth analysis of specific parts of the vehicle running gear, thereby identifying potential fault risks in advance. In addition, the machine learning algorithm used by the system can continuously learn from historical data and optimize the diagnostic model to adapt to changes in the vehicle's operating status and improve the accuracy of fault prediction.

[0130] Furthermore, the system of the present invention can more accurately evaluate the future failure probability of each wheel sub-area and axle sub-area by defining the historical failure probability and using it as a key input parameter. This method can not only improve the reliability of fault prediction, but also, through the preset number of diagnostic steps, the system can flexibly adjust the diagnostic process to adapt to different monitoring needs and conditions. This flexibility and adaptability enables the system to work effectively in a variety of vehicle operating environments, and can provide strong technical support for vehicle maintenance and fault prevention.

[0131] like Figure 2 As shown, in some embodiments, a vehicle running gear on-board fault online monitoring and diagnosis device 200 includes:

[0132] Wheel area division module 201; axle area division module 202; historical fault data acquisition module 203; historical operation status data acquisition module 204; historical fault probability definition module 205; diagnosis model acquisition module 206; future fault probability diagnosis module 207.

[0133] Specifically, the wheel region division module 201 is used to divide each wheel into regions respectively, and each wheel forms at least two wheel sub-regions;

[0134] An axle region division module 202, configured to divide the axle into regions to form at least two axle sub-regions;

[0135] A historical fault data acquisition module 203 is used to acquire a number of historical fault data of each wheel sub-region and each axle sub-region based on a preset time interval;

[0136] A historical operation status data acquisition module 204, configured to acquire a plurality of historical operation status data of each wheel sub-region and each axle sub-region based on the preset time interval;

[0137] A historical fault probability definition module 205, configured to obtain a ratio of the historical fault data to the historical operating status data based on the same wheel sub-region or the same axle sub-region, and define the ratio as a historical fault probability;

[0138] A diagnostic model acquisition module 206 is used to learn and train the historical failure probability of each wheel sub-region and each axle sub-region through a machine learning algorithm to obtain a diagnostic model;

[0139] The future fault probability diagnosis module 207 is used to diagnose the future fault probability of each wheel sub-region and each axle sub-region respectively through the diagnosis model based on a preset number of diagnosis steps.

[0140] It is understandable that the modules recorded in the vehicle running gear on-board fault online monitoring and diagnosis device 200 are the same as those in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the vehicle running gear on-board fault online monitoring and diagnosis system are also applicable to the vehicle running gear on-board fault online monitoring and diagnosis device 200 and the modules contained therein, and will not be repeated here.

[0141] Reference below Figure 3 , which shows a schematic diagram of a structure 300 of an electronic device suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0142] like Figure 3As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 to a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0143] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0144] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0145] The above descriptions are only some preferred embodiments of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A vehicle running gear on-board fault online monitoring and diagnosis system, wherein the vehicle running gear comprises wheels fixedly mounted on the vehicle chassis and axles rotatably mounted on the wheels, characterized in that: The fault online monitoring and diagnosis system performs the following steps: Dividing each wheel into regions respectively, so that each wheel forms at least two wheel sub-regions; Dividing the axle into regions to form at least two axle sub-regions; Acquire a number of historical fault data of each wheel sub-area and each axle sub-area based on a preset time interval; Based on the preset time interval, respectively obtain a number of historical operating status data of each wheel sub-region and each axle sub-region; Acquire a ratio of the historical fault data to the historical operating state data based on the same wheel sub-region or the same axle sub-region, and define the ratio as a historical fault probability; The historical failure probability of each wheel sub-area and each axle sub-area is learned and trained by a machine learning algorithm to obtain a diagnosis model; The future failure probability of each wheel sub-region and each axle sub-region is diagnosed respectively through the diagnosis model based on a preset number of diagnosis steps.

2. The fault online monitoring and diagnosis system according to claim 1, characterized in that: Based on the preset time interval, several historical fault data of each wheel sub-area and each axle sub-area are obtained respectively, including: Monitor the running status of the vehicle running gear in real time through sensors; Obtaining abnormal indicators of the operating status through a data analysis algorithm; Determine whether each wheel sub-region or each axle sub-region has the abnormal indicator; If yes, then record the fault data once for each wheel sub-area or axle sub-area having the abnormal indicator; All historical fault data of each wheel sub-region and each axle sub-region are acquired respectively based on the preset time interval.

3. The fault online monitoring and diagnosis system according to claim 2 is characterized in that: The abnormal indicators of the operating state are obtained by a data analysis algorithm, including: Acquiring operating parameters of the vehicle running gear by sampling at a first preset frequency; Predicting a preset number of fault characteristic values ​​based on each sampling parameter; Defining the fault characteristic value to have the highest confidence level; Obtain the fault feature value with the highest confidence level among all confidence levels as the key fault feature; Calculating the correlation between the key fault characteristic value and the fault characteristic value; The key fault characteristic values ​​with a correlation greater than or equal to a preset ratio are retained as abnormal indicators; Gets a collection of all anomaly indicators.

4. The fault online monitoring and diagnosis system according to claim 1, characterized in that: Based on the preset time interval, a plurality of historical operating status data of each wheel sub-region and each axle sub-region are respectively obtained, including: Acquiring the running status data of the vehicle running gear through continuous monitoring; Acquire fault feature points in the operating status data through a status recognition algorithm; Determine whether each wheel sub-region or each axle sub-region has the fault characteristic point; If so, the wheel sub-region or the axle sub-region where the fault characteristic point exists respectively records the running status data once; All historical operating status data of each wheel sub-area and each axle sub-area are respectively acquired based on the preset time interval.

5. The fault online monitoring and diagnosis system according to claim 4 is characterized in that: Acquiring fault feature points in the operating status data through a status recognition algorithm includes: Standardizing the operating status data to obtain standardized status data; Use key operating parameter analysis technology to identify potential fault feature points; Determine the persistence and stability of fault feature points through time series analysis method; Integrate the confirmed fault feature points to form a fault feature sequence.

6. The fault online monitoring and diagnosis system according to claim 1, characterized in that: The historical failure probabilities of each wheel sub-area and each axle sub-area are learned and trained by machine learning algorithms to obtain a diagnostic model, including: Generate a number of acquisition time points based on a preset time interval; Based on all acquisition time points, the historical failure probability of each wheel sub-area and each axle sub-area is obtained to generate a data set; All datasets were integrated to form a total dataset; performing standardization on the total data set; Dividing the standardized data set into a training set and a validation set according to a preset ratio; Define the topology of the neural network model; The training set is input into the neural network model for iterative training to obtain the diagnostic model corresponding to the minimum root mean square error.

7. The fault online monitoring and diagnosis system according to claim 6, characterized in that: The neural network model characterizes the specific calculation process of the failure probability through the following formula: in, represents the failure probability of the output; Indicates The weight of the input features to the hidden layer; Indicates Input feature values; represents the bias of the hidden layer; Represents the activation function.

8. A vehicle running gear on-board fault online monitoring and diagnosis device, characterized in that: include: A wheel region division module, used for dividing each wheel into regions, each wheel forming at least two wheel sub-regions; An axle region division module, used for dividing the axle into regions to form at least two axle sub-regions; A historical fault data acquisition module, used to acquire a number of historical fault data of each wheel sub-region and each axle sub-region based on a preset time interval; A historical operation status data acquisition module, used for acquiring a plurality of historical operation status data of each wheel sub-region and each axle sub-region based on the preset time interval; A historical fault probability definition module, used for obtaining a ratio of the historical fault data to the historical operating status data based on the same wheel sub-region or the same axle sub-region, and defining the ratio as a historical fault probability; A diagnostic model acquisition module, used to learn and train the historical failure probability of each wheel sub-region and each axle sub-region through a machine learning algorithm to obtain a diagnostic model; The future fault probability diagnosis module is used to diagnose the future fault probability of each wheel sub-region and each axle sub-region respectively through the diagnosis model based on a preset number of diagnosis steps.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores program instructions for implementing the fault online monitoring and diagnosis system as claimed in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores program instructions, and when the program instructions are executed by the processor, the online fault monitoring and diagnosis system according to any one of claims 1 to 7 can be implemented.