Locomotive fault detection method and device and electronic equipment

Abstract

The invention provides a locomotive fault detection method and device and electronic equipment, and relates to the field of data processing. In the method, point cloud data and time-varying operation data of a to-be-detected area corresponding to a target locomotive are obtained, and the time-varying operation data comprise electrical performance data and thermal performance data; constructing a system model corresponding to the to-be-detected area according to the point cloud data and the time-varying operation data; acquiring real-time maintenance data of the to-be-detected area of the target locomotive in the maintenance state; inputting the real-time maintenance data into the system model to detect the operation state of the to-be-detected area, generating a detection result, and displaying the detection result through the system model; if it is determined that the detection result indicates that the potential fault exists in the to-be-detected area, an early warning result is generated, and the early warning result is used for prompting a detection person corresponding to the target locomotive through AR glasses worn by the detection person. By implementing the technical scheme provided by the invention, the fault of the locomotive can be conveniently detected.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a locomotive fault detection method, device and electronic equipment. Background Art

[0002] In the current railway locomotive operation scenario, the stability and reliability of the locomotive are key factors in ensuring the safe and continuous operation of the entire railway system. However, within the current technical framework, the locomotive fault detection mechanism mainly relies on traditional means of manual inspection, which has significant limitations in efficiency and accuracy, and poses a risk to the safe operation of the locomotive that cannot be ignored.

[0003] The limitations of the manual inspection model are specifically reflected in the following aspects: First, this method requires a lot of human resources and material costs, and the inspection cycle is long, making it difficult to achieve real-time monitoring and feedback of the locomotive's operating status; second, the accuracy of the inspection results is highly dependent on the personal experience and skill mastery of the inspectors, which not only increases the subjectivity of the judgment, but also increases the risk of misjudgment; third, the early symptoms of locomotive failures are often mild and difficult to detect, and manual inspections often find it difficult to capture these subtle change signals, thus missing the best time to prevent and respond to failures. In summary, the traditional model of manual inspection is not conducive to detecting locomotive failures.

[0004] Therefore, there is an urgent need for a locomotive fault detection method, device and electronic equipment. Summary of the invention

[0005] The present application provides a locomotive fault detection method, device and electronic equipment, which are convenient for detecting locomotive faults.

[0006] In a first aspect of the present application, a locomotive fault detection method is provided, the method comprising: obtaining point cloud data and time-varying operation data of a target locomotive corresponding to an area to be detected, the time-varying operation data comprising electrical performance data and thermal performance data; constructing a system model corresponding to the area to be detected based on the point cloud data and the time-varying operation data; obtaining real-time maintenance data of the area to be detected when the target locomotive is in a maintenance state; inputting the real-time maintenance data into the system model to detect the operating state of the area to be detected and generate a detection result, which is displayed through the system model; if it is determined that the detection result indicates that there is a potential fault in the area to be detected, generating an early warning result, and the early warning result is used to prompt the detection personnel corresponding to the target locomotive through the AR glasses worn by the detection personnel.

[0007] By adopting the above technical solution, by acquiring the point cloud data and time-varying operation data of the target locomotive, the system can understand the operation status of the locomotive in an all-round and multi-dimensional manner. The point cloud data provides accurate three-dimensional information of the locomotive's physical structure, while the time-varying operation data reflects the dynamic performance changes of the locomotive during operation. This data collection method is more comprehensive and accurate than traditional manual inspections, and can capture more subtle fault signals, thereby improving the accuracy of fault detection. The system model corresponding to the area to be inspected is constructed using the collected data. This model can simulate and reflect the actual state of the locomotive in actual operation. The system model can not only help the inspection personnel better understand the operation status of the locomotive, but also provide strong support for subsequent fault analysis and processing. By obtaining the real-time maintenance data of the target locomotive in the maintenance state and inputting it into the system model, the system can perform real-time detection of the locomotive's operation status. This real-time detection method can timely discover and handle potential faults, thereby avoiding further deterioration of the fault and ensuring the safe operation of the locomotive. The detection results are displayed through the system model, so that the inspection personnel can intuitively understand the operation status and potential faults of the locomotive. This visual display method not only improves the detection efficiency, but also reduces the risk of misjudgment. If it is determined that the test results indicate that there is a potential fault in the area to be detected, an early warning result is generated and prompted through the AR glasses worn by the inspector. This instant prompt method can ensure that the inspector obtains fault information in a timely manner and takes countermeasures quickly, thereby effectively preventing the occurrence or expansion of the fault. The entire detection system realizes the full intelligence and automation of data collection, model construction, real-time detection to early warning prompts. This intelligent and automated detection method not only improves the detection efficiency, but also reduces labor costs, providing a strong guarantee for the safe and efficient operation of the railway system. Therefore, the above method is convenient for detecting locomotive faults.

[0008] Optionally, the acquiring of point cloud data and time-varying operation data of the area to be detected corresponding to the target locomotive specifically includes: receiving original scanning data for the area to be detected sent by the AR glasses; performing optical calculation and physical property statistics on the original scanning data to obtain the point cloud data, the optical calculation includes geometric shape calculation and texture calculation, and the physical property statistics includes size statistics, weight statistics and connection method statistics; receiving voltage load data, current load data and temperature load data for the area to be detected sent by sensor equipment; determining the time-varying operation data according to the voltage load data, current load data and temperature load data of the target locomotive.

[0009] By adopting the above technical solution, the system can obtain detailed three-dimensional information of the area to be detected by receiving the original scanning data sent by AR glasses, including optical properties such as geometry and texture, as well as physical properties such as size, weight, and connection method. This information provides a solid foundation for subsequent modeling and analysis. At the same time, the system also receives voltage load data, current load data, and temperature load data sent by sensor equipment. These time-varying operation data reflect the dynamic performance changes of the locomotive during operation and are crucial for detecting potential faults. Optical calculation and physical property statistics are performed on the original scanning data to obtain accurate point cloud data. Optical calculation includes geometry calculation and texture calculation, which can accurately restore the three-dimensional morphology and surface texture of the area to be detected; physical property statistics provide detailed information about size, weight, and connection method, which helps to understand the physical structure and function of the locomotive. The determination process of time-varying operation data also reflects accuracy. The system comprehensively evaluates the electrical and thermal performance of the locomotive based on voltage load data, current load data, and temperature load data, thereby obtaining accurate time-varying operation data. This process realizes the fusion of optical data and physical data, as well as the combination of time-varying operation data and static structure data. This data fusion method can more comprehensively reflect the operating status of the locomotive and improve the accuracy and reliability of fault detection. Inspectors do not need to manually collect and process data. They only need to wear AR glasses and operate sensor equipment to obtain comprehensive test data. This efficient data acquisition and processing method enables inspectors to discover potential faults more quickly and take corresponding countermeasures to ensure the safe operation of the locomotive.

[0010] Optionally, constructing a system model corresponding to the area to be inspected based on the point cloud data and the time-varying operation data specifically includes: modeling the point cloud data by a preset modeling method, and performing denoising, alignment and splicing processing to obtain a locomotive structure 3D model; adding the voltage load data, the current load data and the temperature load data to the locomotive structure 3D model to obtain the system model.

[0011] By adopting the above technical solution, the point cloud data is modeled by a preset modeling method, and a 3D model of the locomotive structure can be generated. This three-dimensional visualization model can intuitively display the physical structure and spatial layout of the locomotive, so that the inspectors can more clearly understand the overall structure of the locomotive and the relationship between the various components. The 3D model not only improves the intuitiveness of the inspection, but also helps the inspectors to more accurately locate the potential fault area, thereby improving the inspection efficiency. In the modeling process, denoising, registration and splicing of the point cloud data are necessary steps. Denoising can remove noise points in the data and improve the accuracy of the model; registration can align the point cloud data from different perspectives to ensure the integrity of the model; splicing can merge multiple point cloud data into a complete 3D model. These processing steps can ensure that the generated 3D model is consistent with the actual locomotive in terms of geometry and size, providing an accurate basis for subsequent analysis and inspection. By adding voltage load data, current load data and temperature load data to the 3D model of the locomotive structure, a system model containing electrical performance and thermal performance information can be constructed. This multi-source data fusion method can more comprehensively reflect the operating status of the locomotive. The system model not only contains the physical structure information of the locomotive, but also contains the electrical and thermal performance data, which enables inspectors to have a deeper understanding of the performance status of the locomotive and discover potential electrical and thermal faults. The system model provides a convenient tool for fault detection. Inspectors can intuitively view the status of each component in the model and determine whether there is an abnormality based on the electrical and thermal performance data. In addition, the system model can also support simulation analysis and prediction functions to help inspectors predict the development trend of potential faults and formulate corresponding maintenance plans. The system model provides strong support for maintenance decisions. Inspectors can determine the priority and plan of maintenance based on the data and analysis results in the model. This data-based decision-making method can reduce the blindness and uncertainty of maintenance and improve maintenance efficiency and accuracy.

[0012] Optionally, the real-time maintenance data includes real-time current, real-time voltage and real-time temperature, and the real-time maintenance data is input into the system model to detect the operating status of the area to be detected and generate a detection result, specifically including: using a preset analysis model to compare the real-time current, the real-time voltage and the real-time temperature with a preset current range, a preset voltage range and a preset temperature range; if it is determined that the real-time current is within the preset current range and the real-time voltage is within the preset voltage range, and the real-time temperature is not within the preset temperature range and shows an upward trend, then the operating status of the area to be detected is determined to be overheating and / or internal damage; the detection result is generated and displayed through the system model, and the detection result is used to indicate that the area to be detected is overheating and / or internal damage.

[0013] By adopting the above technical solution, the system can accurately detect the operating status of the area to be detected by comparing the real-time current, real-time voltage and real-time temperature with the preset range through the preset analysis model. This comparison method is based on clear standards, reduces the errors caused by subjective judgment, and improves the accuracy of detection. When the real-time temperature is not within the preset temperature range and shows an upward trend, the system can quickly identify the potential risk of overheating and / or internal damage. This timely early warning mechanism helps to prevent the further development of the fault and reduce the downtime and maintenance costs caused by the fault. The system model can intuitively display the detection results, so that the inspectors can quickly understand the operating status of the area to be detected. This visual display method improves the efficiency of information transmission and helps the inspectors make decisions quickly. By comparing the real-time maintenance data with the preset range, the system can directly locate the potential area of ​​overheating and / or internal damage. This greatly shortens the troubleshooting time and improves the maintenance efficiency. Based on the analysis results of the real-time maintenance data, the system can support preventive maintenance strategies. The inspectors can formulate maintenance plans in advance based on the detection results to avoid the occurrence of faults and extend the service life of the locomotive. The system relies on real-time data and preset standards for analysis and decision-making, which reflects the data-driven concept. By continuously accumulating and analyzing data, the system can continuously optimize the preset range and analysis model to improve the accuracy and efficiency of detection.

[0014] Optionally, if it is determined that the detection result indicates that there is a potential fault in the area to be detected, a warning result is generated, which specifically includes: determining the type of fault according to the detection result; determining the target detection personnel who processes the detection result based on the correspondence between the fault type and the detection personnel; and sending the warning result to the AR glasses worn by the target detection personnel, the warning result including the detection result and the fault position corresponding to the detection result in the system model.

[0015] By adopting the above technical solution, the fault type is determined according to the test results, which is the first and most important step of fault warning. Accurately identifying the fault type helps the tester quickly understand the nature of the problem and take correct countermeasures. Based on the correspondence between the fault type and the tester, the target tester who handles the test results is determined. This matching mechanism ensures that each fault can be handled by the tester who is best at handling the fault, improving the processing efficiency and accuracy. The early warning results are sent to the AR glasses worn by the target tester, realizing instant communication and information sharing. This method not only ensures the rapid transmission of information, but also enables the tester to obtain fault information at the first time, providing a guarantee for rapid response and processing. The early warning results include the test results and the fault location corresponding to the test results in the system model. This intuitive display method helps the tester quickly locate the fault area, reduces the time of finding the fault, and improves the maintenance efficiency.

[0016] Optionally, the method further includes: generating a target processing strategy based on the correspondence between the warning result and the processing strategy; marking the target processing strategy on the system model, and displaying the marked system model through the AR glasses.

[0017] By adopting the above technical solution, the system can quickly generate a target processing strategy by directly matching the warning results with the processing strategy. This automated decision-making process reduces the time delay of human decision-making and improves the response speed. The target processing strategy is marked on the system model and displayed through AR glasses, providing intuitive guidance for maintenance personnel. Inspection personnel can directly view the system model on site to understand the fault location and the treatment measures to be taken, thereby improving the accuracy and efficiency of maintenance.

[0018] Optionally, the method also includes: obtaining sample data; inputting the sample data into an adaptive feature fusion network for training to obtain a first training result; superimposing and standardizing the first training result and the sample data to obtain a second training result; inputting the second training result into the adaptive feature fusion network for processing to obtain a third training result; superimposing and standardizing the third training result and the second training result until the sample data similarity matrix is ​​output to obtain the preset analysis model, and the sample data similarity matrix satisfies a preset logistic regression condition.

[0019] By adopting the above technical solution, by obtaining sample data as input, the entire model building process is data-based, which ensures the practicality and accuracy of the model. The sample data comes from actual business scenarios, so the model can better reflect the real situation. Using an adaptive feature fusion network, the network can automatically learn and fuse relevant information between different features. This helps to extract key information from the data and improve the performance of the model. Adaptive feature fusion can also process data from different sources and formats, enhancing the versatility and flexibility of the model. Through multiple iterations of training, the training results are superimposed and standardized with the sample data, and then input into the network again for processing. This iterative process helps to gradually optimize the model parameters and improve the prediction accuracy and generalization ability of the model. Standardization is a very important step in data preprocessing. It helps to eliminate the dimensional differences between different features, so that the model can process each feature more fairly. This helps to improve the stability and accuracy of the model. The final output sample data similarity matrix meets the preset logistic regression conditions, which means that the model can not only accurately classify or identify samples in the feature space, but also further explain and verify the output results of the model through logistic regression. This enhances the interpretability and credibility of the model.

[0020] In a second aspect of the present application, a locomotive fault detection device is provided, the detection device comprising an acquisition module and a processing module, wherein the acquisition module is used to acquire point cloud data and time-varying operation data of a target locomotive corresponding to a to-be-detected area, the time-varying operation data comprising electrical performance data and thermal performance data; the processing module is used to construct a system model corresponding to the to-be-detected area based on the point cloud data and the time-varying operation data; the acquisition module is also used to acquire real-time maintenance data of the to-be-detected area when the target locomotive is in a maintenance state; the processing module is also used to input the real-time maintenance data into the system model to detect the operating state of the to-be-detected area and generate a detection result, which is displayed through the system model; the processing module is also used to generate an early warning result if it is determined that the detection result indicates that there is a potential fault in the to-be-detected area, and the early warning result is used to prompt the detection personnel corresponding to the target locomotive through the AR glasses worn by the detection personnel.

[0021] In the third aspect of the present application, an electronic device is provided, which includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device performs the method described above.

[0022] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, the method described above is executed.

[0023] In summary, one or more technical solutions provided in this application have at least the following technical effects or advantages: By acquiring the point cloud data and time-varying operation data of the target locomotive, the system can understand the operation status of the locomotive in an all-round and multi-dimensional way. The point cloud data provides accurate three-dimensional information of the locomotive's physical structure, while the time-varying operation data reflects the dynamic performance changes of the locomotive during operation. This data collection method is more comprehensive and accurate than traditional manual inspections, and can capture more subtle fault signals, thereby improving the accuracy of fault detection. The collected data is used to build a system model corresponding to the area to be inspected, which can simulate and reflect the actual state of the locomotive in actual operation. The system model can not only help inspectors better understand the operation status of the locomotive, but also provide strong support for subsequent fault analysis and processing. By acquiring the real-time maintenance data of the target locomotive in the maintenance state and inputting it into the system model, the system can perform real-time detection of the locomotive's operation status. This real-time detection method can detect and handle potential faults in a timely manner, thereby avoiding further deterioration of the fault and ensuring the safe operation of the locomotive. The detection results are displayed through the system model, allowing inspectors to intuitively understand the operation status and potential faults of the locomotive. This visual display method not only improves the detection efficiency, but also reduces the risk of misjudgment. If it is determined that the test results indicate that there is a potential fault in the area to be tested, an early warning result is generated and prompted through the AR glasses worn by the test personnel. This instant prompt method can ensure that the test personnel obtain fault information in a timely manner and take countermeasures quickly, thereby effectively preventing the occurrence or expansion of the fault. The entire detection system realizes the full intelligence and automation of data collection, model construction, real-time detection to early warning prompts. This intelligent and automated detection method not only improves the detection efficiency, but also reduces labor costs, providing a strong guarantee for the safe and efficient operation of the railway system. Therefore, the above method is convenient for detecting locomotive faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic flow chart of a locomotive fault detection method provided in an embodiment of the present application.

[0025] Figure 2 Another schematic flow chart of a locomotive fault detection method provided in an embodiment of the present application.

[0026] Figure 3 A module schematic diagram of a locomotive fault detection device provided in an embodiment of the present application.

[0027] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0028] Explanation of the reference numerals: 31, acquisition module; 32, processing module; 41, processor; 42, communication bus; 43, user interface; 44, network interface; 45, memory. DETAILED DESCRIPTION

[0029] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0030] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0031] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0032] In the current railway locomotive operation practice, the stability and reliability of locomotives constitute the indispensable cornerstone for the safe and continuous operation of the railway system. However, under the current technical architecture, the detection of locomotive faults still mainly relies on the traditional method of manual inspection, which has obvious limitations in efficiency and accuracy, and poses a potential threat to the safe operation of locomotives.

[0033] The limitations of manual inspection are as follows: first, it requires a lot of human resources and material investment, and due to the long inspection cycle, it is difficult to monitor and feedback the running status of the locomotive in a timely manner; second, the accuracy of the inspection results depends largely on the personal experience and skill level of the inspectors, which not only increases the subjectivity of the results, but also increases the possibility of misjudgment; finally, the early stage of locomotive failures often manifests as mild and difficult to detect symptoms, and manual inspections often find it difficult to capture these subtle changes, thus missing the best time to prevent and deal with failures. Therefore, the traditional manual inspection model has obvious shortcomings in locomotive fault detection.

[0034] In order to solve the above technical problems, the present application provides a locomotive fault detection method, referring to Figure 1 , Figure 1 A flow chart of a locomotive fault detection method provided in an embodiment of the present application. The method is applied to a server and specifically includes steps S110 to S150, which are as follows: S110, obtaining point cloud data and time-varying operation data of the target locomotive corresponding to the area to be detected, where the time-varying operation data includes electrical performance data and thermal performance data.

[0035] Specifically, the server is the central processing unit responsible for the overall locomotive fault detection task, which can be a computer system or a server cluster. Point cloud data is a collection of a large number of three-dimensional coordinate points, which may also contain other information such as color and intensity. These points together describe the three-dimensional shape and appearance of an object or scene. In locomotive detection, point cloud data can come from devices such as laser radar and three-dimensional scanners to obtain the three-dimensional shape information of the area to be detected, including key components of the locomotive, such as wheels, bogies, and car bodies. For example, assuming that the target locomotive is a high-speed train, the server needs to obtain the point cloud data of its wheels. At this time, a laser radar installed beside the track can be used to scan the wheels when the train passes by to generate point cloud data of the wheels. These data contain the three-dimensional shape information of the wheels and can be used to detect defects such as wear and cracks on the wheels.

[0036] Time-varying operation data refers to locomotive operation data that changes over time. These data include electrical performance data and thermal performance data. Electrical performance data mainly reflects the state of the locomotive's electrical system, such as current, voltage, power, etc.; thermal performance data mainly reflects the thermal state of the locomotive, such as temperature, heat flow, etc. These data are of great significance for evaluating the locomotive's operating status and predicting potential faults.

[0037] For example, suppose the target locomotive is running and the server needs to obtain the current data of its traction motor. At this time, the current data can be collected in real time through the current sensor installed on the motor and transmitted to the server for analysis. If the current data is abnormal, such as too large, too small, or fluctuates too much, it may indicate that the motor is faulty or needs maintenance. Taking the target locomotive as an example, the server needs to obtain the temperature data of its braking system. At this time, temperature sensors can be installed at key parts of the braking system to collect temperature data in real time and transmit it to the server. If the temperature data rises abnormally, it may indicate that the braking system is overheating and immediate measures need to be taken to prevent failures.

[0038] In a possible implementation, point cloud data and time-varying operation data of the target locomotive corresponding to the area to be detected are obtained, specifically including: receiving original scanning data for the area to be detected sent by AR glasses; performing optical calculation and physical property statistics on the original scanning data to obtain point cloud data, the optical calculation includes geometric shape calculation and texture calculation, and the physical property statistics include size statistics, weight statistics and connection method statistics; receiving voltage load data, current load data and temperature load data for the area to be detected sent by sensor equipment; determining the time-varying operation data according to the voltage load data, current load data and temperature load data of the target locomotive.

[0039] Specifically, the server first receives the original scan data for the area to be detected sent by the AR glasses. These data are in the original form of images, videos or three-dimensional point cloud data, which contain the basic shape and texture information of the area to be detected. The original scan data received by the server needs to be further processed to obtain point cloud data. This includes two parts: optical calculation and physical property statistics. Optical calculation mainly involves the processing of image or video data to extract the geometric shape and texture information of the area to be detected. Geometric shape calculation may involve algorithms such as edge detection and contour extraction to obtain the three-dimensional shape of the object; texture calculation may involve the analysis of features such as color, brightness, and contrast to obtain the surface texture information of the object. Physical property statistics is a more in-depth physical property analysis of the scan data, such as size statistics, that is, measuring the length, width, height and other size information of the object, such as weight statistics, that is, estimating through shape and material through AR glasses, and connection method statistics, that is, analyzing the connection method and structural relationship between the various parts of the object.

[0040] The server also receives voltage load data, current load data, and temperature load data for the area to be detected from sensor devices such as voltage sensors, current sensors, and temperature sensors. These data are key indicators for real-time detection of the locomotive's operating status and can reflect the locomotive's electrical and thermal performance. The received sensor data needs to be further processed and analyzed to determine the locomotive's time-varying operating data. This involves operations such as data cleaning, screening, conversion, and aggregation to extract useful information and generate formatted data that can be used for subsequent analysis and decision-making.

[0041] S120: construct a system model corresponding to the area to be detected based on the point cloud data and the time-varying operation data.

[0042] Specifically, the server will first integrate the acquired point cloud data and time-varying operation data. The point cloud data provides the three-dimensional shape, texture and physical property information of the area to be inspected, while the time-varying operation data provides real-time status information such as the electrical performance and thermal performance of the area during operation. Based on the data integration, the server will use these data to build a system model. This system model is a digital representation of the area to be inspected and its operating status, which can be used to simulate, analyze and predict the performance and behavior of the area. The system model contains multiple elements, such as geometric models, which are constructed from point cloud data; physical models, which describe the physical properties and behavior of objects; electrical models, which describe the structure and performance of electrical systems; thermal models, which describe heat transfer and temperature distribution, etc. In the process of building the system model, the server may use various algorithms and technologies, such as three-dimensional reconstruction, physical simulation, electrical network analysis, and thermal conduction analysis. The constructed system model can be used for a variety of applications, such as fault prediction, performance optimization, maintenance decision-making, etc. By simulating and analyzing the model, potential problems can be predicted and measures can be taken in advance to avoid or reduce the occurrence of faults.

[0043] In a possible implementation, a system model corresponding to the area to be inspected is constructed based on the point cloud data and time-varying operation data, specifically including: modeling the point cloud data using a preset modeling method, and performing denoising, alignment and splicing processing to obtain a 3D model of the locomotive structure; adding voltage load data, current load data and temperature load data to the 3D model of the locomotive structure to obtain a system model.

[0044] Specifically, the server will first use one or more preset modeling methods, such as surface reconstruction, voxelization, and meshing, to process the point cloud data to construct a three-dimensional model of the locomotive structure. The choice of these modeling methods depends on the characteristics of the point cloud data, the accuracy requirements of the modeling, and the needs of subsequent applications. During the modeling process, the server also needs to denoise the point cloud data to eliminate noise points in the data and improve the accuracy of the model. Registration processing refers to aligning point cloud data acquired from different perspectives or different time periods to the same coordinate system to ensure the consistency and integrity of the model. Splicing processing is to merge point cloud data from multiple parts into a complete model. This involves complex algorithms and technologies to ensure smoothness and continuity at the splicing. After the above processing steps, the server will obtain a complete three-dimensional model of the locomotive structure. This model not only contains the geometric shape information of the locomotive, but may also contain additional information such as texture and color. Next, the server will add the acquired time-varying operating data such as voltage load data, current load data, and temperature load data to the three-dimensional model of the locomotive structure. These data exist in the form of sensor measurements and need to be properly processed, such as data cleaning, conversion, interpolation, etc., to match the format and accuracy requirements of the model. After adding these data, the 3D model of the locomotive structure will be upgraded to a system model that includes electrical and thermal performance information. This system model can more comprehensively reflect the operating status and performance characteristics of the locomotive.

[0045] For example, assuming that the target locomotive is a freight train, the server needs to build a system model of its wheels and brake system. The wheels and brake system are scanned by AR glasses or a 3D scanner to obtain point cloud data. The server uses a surface reconstruction algorithm to model the point cloud data, and performs denoising, registration and splicing to obtain a complete and accurate 3D model of the wheel and brake system. Sensors are installed at key parts of the circuit and brake system near the wheel to detect operating data such as voltage, current and temperature in real time. The server adds the acquired voltage load data, current load data and temperature load data to the 3D model of the wheel and brake system. For example, the voltage and current data of the circuit near the wheel can be represented on the model in the form of color coding to intuitively display the voltage distribution and current flow of the circuit. At the same time, the temperature data of the brake system can be represented on the model in the form of a heat map to reflect the temperature change and heat distribution during the braking process. The constructed system model can be used for a variety of applications, such as fault prediction, performance evaluation, maintenance decision-making, etc. By analyzing the voltage, current and temperature data in the model, potential faults of the wheel and brake system, such as circuit short circuit and brake failure, can be predicted. Based on the prediction results, targeted maintenance plans can be formulated and measures can be taken in advance to avoid or reduce the occurrence of failures. Through the above process, the server can use point cloud data and time-varying operation data to build a three-dimensional model of the locomotive structure and upgrade it to a system model containing electrical and thermal performance information, providing important support for subsequent fault prediction, performance evaluation and maintenance decisions.

[0046] S130, obtaining real-time maintenance data of the area to be inspected of the target locomotive in a maintenance state.

[0047] Specifically, the target locomotive is in a maintenance state, which means that the target locomotive is currently undergoing regular maintenance, inspection or repair work. In this state, the locomotive will stop running so that technicians can perform detailed inspections and necessary repairs. The area to be inspected refers to the specific parts or components of the locomotive that need to be inspected. This may include key components such as wheels, brake systems, electrical systems, and transmission systems. The server will focus on obtaining real-time maintenance data for these areas. Real-time maintenance data refers to data about the current state of the area to be inspected obtained through sensors, measuring tools, or manual records during the maintenance process. These data may include dimensional measurement, material property testing, electrical parameter measurement, temperature detection, etc. Real-time means that these data are obtained in real time during the maintenance process, rather than through review or analysis afterwards. This helps technicians to understand the current state of the locomotive in a timely manner and make accurate judgments and decisions. The server obtains these real-time maintenance data from sensors, measuring tools, or recording systems connected to the locomotive through specific interfaces or protocols. The server may store, process, and analyze these data for subsequent use.

[0048] S140, inputting the real-time maintenance data into the system model to detect the operating status of the area to be detected, and generating the detection results, which are displayed through the system model.

[0049] Specifically, in this step, the server inputs the previously acquired real-time maintenance data, such as dimensional measurement, material performance detection, electrical parameter measurement, temperature detection, etc., into the previously constructed system model. This system model may be a three-dimensional model based on the locomotive structure, and has integrated time-varying operating data such as voltage, current and temperature. Once the real-time maintenance data is input into the system model, the server can use this data to simulate, analyze and evaluate the operating status of the area to be detected. This may involve complex algorithms and technologies, such as physical simulation, electrical network analysis, heat conduction analysis, etc., to accurately reflect the actual status of the area to be detected. After the above analysis and evaluation, the server will generate a test result. This test result may be one or more numerical values, charts, images or animations, which are used to intuitively display the operating status of the area to be detected. The test results may include but are not limited to information such as the degree of wear, performance degradation, and potential failure points. Finally, the server displays the generated test results through the system model. This means that the test results will be presented to the user in the form of three-dimensional visualization, two-dimensional charts or interactive interfaces. Users can intuitively understand the operating status of the area to be detected through these displays and make decisions accordingly.

[0050] In a possible implementation, the real-time maintenance data includes real-time current, real-time voltage and real-time temperature. The real-time maintenance data is input into the system model to detect the operating status of the area to be detected and generate a detection result, which specifically includes: using a preset analysis model to compare the real-time current, real-time voltage and real-time temperature with a preset current range, a preset voltage range and a preset temperature range; if it is determined that the real-time current is within the preset current range and the real-time voltage is within the preset voltage range, and the real-time temperature is not within the preset temperature range and is on an upward trend, then the operating status of the area to be detected is determined to be overheating and / or internal damage; generate a detection result, and display the detection result through the system model, the detection result is used to indicate that the area to be detected is overheating and / or internal damage.

[0051] Specifically, the server uses a preset analysis model to process the real-time data. This analysis model may be based on a series of algorithms, rules or thresholds to compare the real-time data with preset current ranges, voltage ranges and temperature ranges. The real-time current is compared with the preset current range, the real-time voltage is compared with the preset voltage range, and the real-time temperature is compared with the preset temperature range. The purpose of these comparisons is to determine whether the real-time data is within the preset range, so as to evaluate the operating status of the area to be detected. If the real-time current and real-time voltage are both within the preset range, but the real-time temperature is not within the preset temperature range and is on an upward trend, this indicates that the area to be detected may be overheated or internally damaged. Once the operating status of the area to be detected is determined, such as overheating or internal damage, the server will generate a test result. This test result may include one or more indicators to clearly indicate the problem in the area to be detected. The server displays the generated test results through the system model. This means that the test results will be presented in the system model in some form, such as three-dimensional visualization, two-dimensional charts or text messages, so that users can intuitively see the problems in the area to be detected. The main purpose of the test result is to indicate whether the area to be detected has problems with overheating or internal damage. This helps technicians quickly locate the problem and take appropriate measures to repair it.

[0052] For example, assume that the target locomotive is an electric locomotive, and its traction motor is the area to be detected. During the operation of the locomotive, the sensor continuously detects the real-time current, real-time voltage, and real-time temperature of the traction motor. The server uses a preset analysis model that defines the normal current range, normal voltage range, and normal temperature range of the traction motor. The server compares the real-time current with the preset current range and finds that the real-time current is within the normal range. The server compares the real-time voltage with the preset voltage range and finds that the real-time voltage is also within the normal range. However, when the server compares the real-time temperature with the preset temperature range, it finds that the real-time temperature exceeds the normal range and is still rising. Based on these comparison results, the server determines that the traction motor may be overheated or internally damaged. The server generates a detection result that clearly indicates that the traction motor is overheated or internally damaged. The server displays the detection results through the system model, and the user can see in the system model that the traction motor is marked as overheated or damaged. After seeing the detection results, the technician can take immediate action, such as stopping the machine for inspection, replacing damaged parts, etc., to ensure the safe operation of the locomotive. Through the above process, the server can use real-time maintenance data to detect the operating status of the area to be inspected in the system model, and generate and display the corresponding test results, providing important reference information for technicians to help them quickly locate and solve problems.

[0053] S150. If it is determined that the detection result indicates that there is a potential fault in the area to be detected, a warning result is generated. The warning result is used to prompt the detection personnel corresponding to the target locomotive through the AR glasses worn by the detection personnel.

[0054] Specifically, the server has detected the operating status of the area to be inspected through the previous analysis process and generated the test results. If the test results indicate that there is a potential fault in the area to be inspected, such as overheating, internal damage, performance degradation, etc., the server will proceed to the next step. Once a potential fault is determined, the server will immediately generate an early warning result. This early warning result may contain specific information about the potential fault, such as the type, location, severity, etc. of the fault, as well as recommended countermeasures, such as shutdown inspection, replacement of parts, adjustment of parameters, etc. The server sends the generated early warning results to the AR glasses worn by the inspector. AR glasses are an augmented reality device that can superimpose virtual information on the real world to provide additional visual information to the wearer. When the inspectors wear AR glasses, they can see the prompt information of the early warning results. This information may be presented in the form of text, images, animations or sounds so that the inspectors can quickly understand and respond.

[0055] In a possible implementation, if it is determined that the detection result indicates that there is a potential fault in the area to be detected, a warning result is generated, which specifically includes: determining the type of fault based on the detection result; determining the target detection personnel who processes the detection result based on the correspondence between the fault type and the detection personnel; and sending the warning result to the AR glasses worn by the target detection personnel, the warning result including the detection result and the fault position corresponding to the detection result in the system model.

[0056] Specifically, when the server determines through the previous analysis process that the test result indicates that there is a potential fault in the area to be tested, it will first determine the type of fault based on the specific content of the test result. The type of fault may include overheating, short circuit, wear, looseness, damage, etc., depending on the nature of the area to be tested and the detection capability of the sensor. Once the type of fault is determined, the server will determine the target tester who handles the test result based on the correspondence between the type of fault and the tester. This correspondence may be preset when the system is configured, or it may be dynamically adjusted during actual operation. The correspondence may be based on factors such as the professional skills, experience, geographical location, or current working status of the tester. After the target tester is determined, the server will send an early warning result to the AR glasses they wear. The early warning result includes not only the previously generated test results, but also the corresponding fault locations of these test results in the system model. In this way, the tester can not only understand the specific information of the fault, but also intuitively see the location of the fault in the system model, which helps them locate the problem faster and take corresponding measures.

[0057] In one possible implementation, refer to Figure 2 , Figure 2 Another flow chart of a locomotive fault detection method provided in an embodiment of the present application includes steps S210 to S220, which are as follows: S210, generating a target processing strategy according to the correspondence between the warning result and the processing strategy; S220, marking the target processing strategy on the system model, and displaying the marked system model through AR glasses.

[0058] Specifically, the server stores one or more correspondence tables that define the processing strategies that should be adopted for different warning results. These processing strategies may include shutdown inspection, replacement of parts, adjustment of parameters, repair or replacement of the entire system, etc. The correspondence may be generated based on experience, expert system, machine learning algorithm, etc. When the server receives the warning result, it automatically generates a target processing strategy based on the warning result and the stored correspondence table. This strategy is the most appropriate processing method that the server considers for the current warning result. After the target processing strategy is generated, the server will mark this strategy on the system model in some form. This can be achieved by adding annotations, highlighting, changing color or shape on the model. The purpose of marking is to allow the inspector to intuitively see the processing strategy that should be taken. Finally, the server will send the marked system model to the AR glasses worn by the inspector. In this way, the inspector can see the system model with the processing strategy through the AR glasses, so as to have a clearer understanding of what measures should be taken to solve the problem.

[0059] In one possible implementation, sample data is obtained; the sample data is input into an adaptive feature fusion network for training to obtain a first training result; the first training result is superimposed and standardized with the sample data to obtain a second training result; the second training result is input into an adaptive feature fusion network for processing to obtain a third training result; the third training result is superimposed and standardized with the second training result until a sample data similarity matrix is ​​output to obtain a preset analysis model, and the sample data similarity matrix satisfies a preset logistic regression condition.

[0060] Specifically, before training the model, a set of sample data needs to be collected first. These data should be able to represent various situations of the problem to be analyzed so that the model can learn the characteristics and laws therein. The collected sample data is input into the adaptive feature fusion network. This network may be a deep learning model that can automatically learn the characteristics of the data and fuse them, thereby extracting information useful for solving the problem. Through training, the network will output a preliminary result, namely the first training result. This result may not be very accurate, but it already contains some information learned from the sample data. The first training result is superimposed with the original sample data. The purpose of superposition is to compare the learning results of the network with the original data for subsequent processing. Then standardization is performed to eliminate the influence of dimensional differences and outliers in the data on the results. Standardization usually includes converting the data into a distribution with a mean of 0 and a standard deviation of 1. The superimposed and standardized data, namely the second training result, is input into the adaptive feature fusion network again for processing. The network will continue to learn according to the new input and output a more accurate result, namely the third training result. The third training result is superimposed and standardized with the second training result, and then input into the network again for processing. This process will be repeated many times, and each time a new training result will be output, superimposed and standardized. Through continuous iterations, the network's learning ability will gradually increase, and the output results will become closer and closer to the actual situation. When the iteration reaches a certain number of times, or the output result meets a certain preset condition, such as the preset logistic regression condition, the training process will stop. At this time, the network will output a sample data similarity matrix. This matrix represents the similarity between sample data and can be used for subsequent analysis and prediction. The model trained through the above process is the preset analysis model. This model has learned the characteristics and rules of the sample data and can be used to analyze and predict new data.

[0061] The present application also provides a locomotive fault detection device, referring to Figure 3 , Figure 3A module schematic diagram of a locomotive fault detection device provided in an embodiment of the present application. The locomotive detection device is a server, and the server includes an acquisition module 31 and a processing module 32, wherein the acquisition module 31 acquires point cloud data and time-varying operation data of the target locomotive corresponding to the area to be detected, and the time-varying operation data includes electrical performance data and thermal performance data; the processing module 32 constructs a system model corresponding to the area to be detected based on the point cloud data and the time-varying operation data; the acquisition module 31 acquires real-time maintenance data of the area to be detected in the maintenance state of the target locomotive; the processing module 32 inputs the real-time maintenance data into the system model to detect the operating state of the area to be detected, and generates a detection result, which is displayed through the system model; if the processing module 32 determines that the detection result indicates that there is a potential fault in the area to be detected, it generates an early warning result, and the early warning result is used to prompt the detection personnel corresponding to the target locomotive through the AR glasses worn by the detection personnel.

[0062] In a possible implementation, the acquisition module 31 acquires point cloud data and time-varying operation data of the target locomotive corresponding to the area to be detected, specifically including: the acquisition module 31 receives the original scanning data for the area to be detected sent by the AR glasses; the processing module 32 performs optical calculations and physical property statistics on the original scanning data to obtain point cloud data, the optical calculations include geometric shape calculations and texture calculations, and the physical property statistics include size statistics, weight statistics, and connection method statistics; the acquisition module 31 receives the voltage load data, current load data, and temperature load data for the area to be detected sent by the sensor device; the processing module 32 determines the time-varying operation data based on the voltage load data, current load data, and temperature load data of the target locomotive.

[0063] In a possible implementation, the processing module 32 constructs a system model corresponding to the area to be inspected based on the point cloud data and the time-varying operation data, specifically including: the processing module 32 models the point cloud data using a preset modeling method, and performs denoising, alignment and splicing processing to obtain a 3D model of the locomotive structure; the processing module 32 adds the voltage load data, current load data and temperature load data to the 3D model of the locomotive structure to obtain a system model.

[0064] In a possible implementation, the real-time maintenance data includes real-time current, real-time voltage and real-time temperature. The processing module 32 inputs the real-time maintenance data into the system model to detect the operating status of the area to be detected and generate a detection result, specifically including: the processing module 32 uses a preset analysis model to compare the real-time current, real-time voltage and real-time temperature with the preset current range, preset voltage range and preset temperature range; if the processing module 32 determines that the real-time current is within the preset current range and the real-time voltage is within the preset voltage range, and the real-time temperature is not within the preset temperature range and is on an upward trend, then it is determined that the operating status of the area to be detected is overheating and / or internal damage; the processing module 32 generates a detection result and displays the detection result through the system model, and the detection result is used to indicate that the area to be detected is overheating and / or internal damage.

[0065] In a possible implementation, if the processing module 32 determines that the detection result indicates that there is a potential fault in the area to be detected, it generates a warning result, which specifically includes: the processing module 32 determines the type of fault according to the detection result; the processing module 32 determines the target detection personnel who processes the detection result based on the correspondence between the fault type and the detection personnel; the processing module 32 sends the warning result to the AR glasses worn by the target detection personnel, and the warning result includes the detection result and the fault position corresponding to the detection result in the system model.

[0066] In a possible implementation, the processing module 32 generates a target processing strategy based on the correspondence between the warning result and the processing strategy; the processing module 32 marks the target processing strategy on the system model, and displays the marked system model through AR glasses.

[0067] In one possible implementation, the acquisition module 31 acquires sample data; the processing module 32 inputs the sample data into an adaptive feature fusion network for training to obtain a first training result; the processing module 32 superimposes and standardizes the first training result and the sample data to obtain a second training result; the processing module 32 inputs the second training result into an adaptive feature fusion network for processing to obtain a third training result; the processing module 32 superimposes and standardizes the third training result and the second training result until a sample data similarity matrix is ​​output to obtain a preset analysis model, and the sample data similarity matrix satisfies a preset logistic regression condition.

[0068] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0069] The present application also provides an electronic device, referring to Figure 4 , Figure 4 The electronic device may include: at least one processor 41 , at least one network interface 44 , a user interface 43 , a memory 45 , and at least one communication bus 42 .

[0070] The communication bus 42 is used to realize the connection and communication between these components.

[0071] The user interface 43 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 43 may also include a standard wired interface and a wireless interface.

[0072] The network interface 44 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0073] Among them, the processor 41 may include one or more processing cores. The processor 41 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 45, and calling data stored in the memory 45. Optionally, the processor 41 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 41 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 41, and it can be implemented separately through a chip.

[0074] Among them, the memory 45 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 45 includes a non-transitory computer-readable storage medium. The memory 45 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 45 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 45 may also be optionally at least one storage device located away from the aforementioned processor 41. As Figure 4 As shown, the memory 45 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of a locomotive fault detection method.

[0075] exist Figure 4 In the electronic device shown, the user interface 43 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 41 can be used to call an application program for a locomotive fault detection method stored in the memory 45. When executed by one or more processors, the electronic device executes one or more methods in the above-mentioned embodiments.

[0076] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0077] The present application also provides a computer-readable storage medium, which stores instructions. When executed by one or more processors, the electronic device executes one or more of the methods described in the above embodiments.

[0078] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0079] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0080] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0081] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0082] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0083] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, it will be easy for those skilled in the art to think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A locomotive fault detection method, characterized in that: The method comprises: Acquire point cloud data and time-varying operation data of the target locomotive corresponding to the area to be detected, wherein the time-varying operation data includes electrical performance data and thermal performance data; Constructing a system model corresponding to the area to be detected according to the point cloud data and the time-varying operation data; Acquire real-time maintenance data of the area to be inspected when the target locomotive is in maintenance state; Inputting the real-time maintenance data into the system model to detect the operating status of the area to be detected and generate a detection result, which is displayed through the system model; If it is determined that the detection result indicates that there is a potential fault in the area to be detected, an early warning result is generated, and the early warning result is used to prompt the detection personnel corresponding to the target locomotive through the AR glasses worn by the detection personnel.

2. The locomotive fault detection method according to claim 1, characterized in that: The step of obtaining the point cloud data and time-varying operation data of the target locomotive corresponding to the area to be detected specifically includes: Receiving raw scan data for the area to be detected sent by the AR glasses; Performing optical calculation and physical property statistics on the original scan data to obtain the point cloud data, wherein the optical calculation includes geometric shape calculation and texture calculation, and the physical property statistics includes size statistics, weight statistics, and connection mode statistics; Receiving voltage load data, current load data and temperature load data for the area to be detected sent by the sensor device; The time-varying operation data is determined according to the voltage load data, the current load data and the temperature load data of the target locomotive.

3. The locomotive fault detection method according to claim 2, characterized in that: The constructing a system model corresponding to the area to be detected according to the point cloud data and the time-varying operation data specifically includes: Modeling the point cloud data by a preset modeling method, and performing denoising, registration and splicing processing to obtain a 3D model of the locomotive structure; The voltage load data, the current load data and the temperature load data are added to the locomotive structure 3D model to obtain the system model.

4. The locomotive fault detection method according to claim 2, characterized in that: The real-time maintenance data includes real-time current, real-time voltage and real-time temperature. The real-time maintenance data is input into the system model to detect the operating status of the area to be detected and generate a detection result, which specifically includes: Using a preset analysis model to compare the real-time current, the real-time voltage and the real-time temperature with a preset current range, a preset voltage range and a preset temperature range; If it is determined that the real-time current is within the preset current range and the real-time voltage is within the preset voltage range, and the real-time temperature is not within the preset temperature range and is on an upward trend, then it is determined that the operating state of the area to be detected is overheating and / or internal damage; The detection result is generated and displayed through the system model, wherein the detection result is used to indicate overheating and / or internal damage of the area to be detected.

5. The locomotive fault detection method according to claim 1, characterized in that: If it is determined that the detection result indicates that there is a potential fault in the area to be detected, then generating a warning result specifically includes: Determine the fault type according to the detection result; Determine a target inspection person who will handle the inspection result based on the correspondence between the fault type and the inspection person; The early warning result is sent to the AR glasses worn by the target detection person, wherein the early warning result includes the detection result and the fault location corresponding to the detection result in the system model.

6. The locomotive fault detection method according to claim 1, characterized in that: The method further comprises: Generate target processing strategies based on the corresponding relationship between warning results and processing strategies; The target processing strategy is marked on the system model, and the marked system model is displayed through the AR glasses.

7. The locomotive fault detection method according to claim 4, characterized in that: The method further comprises: Get sample data; Inputting the sample data into an adaptive feature fusion network for training to obtain a first training result; After superimposing and normalizing the first training result and the sample data, a second training result is obtained; Inputting the second training result into the adaptive feature fusion network for processing to obtain a third training result; The third training result and the second training result are superimposed and standardized until the sample data similarity matrix is ​​output to obtain the preset analysis model, and the sample data similarity matrix satisfies the preset logistic regression condition.

8. A locomotive fault detection device, characterized in that: The detection device comprises an acquisition module (31) and a processing module (32), wherein: The acquisition module (31) is used to acquire point cloud data and time-varying operation data of the target locomotive corresponding to the area to be detected, wherein the time-varying operation data includes electrical performance data and thermal performance data; The processing module (32) is used to construct a system model corresponding to the area to be detected based on the point cloud data and the time-varying operation data; The acquisition module (31) is also used to acquire real-time maintenance data of the area to be detected when the target locomotive is in a maintenance state; The processing module (32) is further used to input the real-time maintenance data into the system model to detect the operating status of the area to be detected and generate a detection result, and the detection result is displayed through the system model; The processing module (32) is further configured to generate an early warning result if it is determined that the detection result indicates that there is a potential fault in the area to be detected, and the early warning result is used to prompt a detection personnel corresponding to the target locomotive through the AR glasses worn by the detection personnel.

9. An electronic device, characterized in that: The electronic device comprises a processor (41), a memory (45), a user interface (43) and a network interface (44), wherein the memory (45) is used to store instructions, the user interface (43) and the network interface (44) are both used to communicate with other devices, and the processor (41) is used to execute the instructions stored in the memory (45) so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images