New energy automobile electrical equipment detection system and method

Abstract

The invention provides a new energy automobile electrical equipment detection system and method. The system comprises an intelligent sensing module used for collecting multi-dimensional electrical data of an automobile in real time; the data processing module is used for performing standardization processing on the multi-dimensional electrical data to generate a standardized electrical data set; the AI diagnosis module is used for analyzing the standardized electrical data based on an AI fault prediction and diagnosis model, determining health states of various electrical devices on the automobile for evaluation and prediction, and generating a diagnosis report; and the remote interaction module is used for the maintenance personnel to access and obtain the diagnosis report based on the terminal equipment and send fault processing guidance to the automobile user. The vehicle fault is diagnosed and predicted through the AI model, so that the vehicle diagnosis efficiency is greatly improved, the troubleshooting time is greatly shortened, the loss and inconvenience brought to a user by equipment shutdown are reduced, meanwhile, remote maintenance guidance is provided, the space and time limitation of vehicle maintenance is broken through, and the trouble and unease brought by the fault of the user are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent detection of automotive electrical equipment, and particularly relates to a detection system and method for electrical equipment of new energy vehicles. Background Art

[0002] With the popularization of new energy vehicles, the reliability of their electrical equipment (such as battery management systems, motor controllers, charging and discharging modules, etc.) directly affects vehicle performance and safety. Traditional detection methods rely on manual experience or offline detection equipment, and have problems such as poor real-time performance, low diagnostic efficiency, and high maintenance costs. In the prior art, some systems attempt to introduce sensors for data acquisition, but lack intelligent prediction and remote maintenance capabilities, and are difficult to meet the high-efficiency detection requirements under complex working conditions. Therefore, the present invention proposes a detection system and method for electrical equipment of new energy vehicles. Summary of the Invention

[0003] The present invention provides a detection system and method for electrical equipment of new energy vehicles to solve the above problems, and proposes an integrated system for real-time vehicle status monitoring, intelligent fault diagnosis and remote maintenance. Based on the AI fault prediction and diagnosis model, it intelligently detects vehicle faults, greatly improves the vehicle diagnosis efficiency, predicts potential vehicle faults, significantly shortens the fault troubleshooting time, minimizes the losses and inconveniences brought to users by equipment downtime, and at the same time provides remote repair guidance for users, breaking the space and time limitations of vehicle repair and reducing the troubles and uneasiness brought to users by faults.

[0004] The present invention provides a detection system for electrical equipment of new energy vehicles, including:

[0005] An intelligent sensing module, configured to collect multi-dimensional electrical data of the vehicle in real time;

[0006] A data processing module, configured to perform standardization processing on the multi-dimensional electrical data to generate a standardized electrical data set;

[0007] An AI diagnosis module, configured to analyze the standardized electrical data based on the AI fault prediction and diagnosis model, evaluate and predict the health status of various electrical equipment on the vehicle, and generate a diagnostic report;

[0008] A remote interaction module, configured to enable maintenance personnel to access the diagnostic report based on a terminal device and send fault handling guidance to vehicle users.

[0009] Preferably, in a detection system for electrical equipment of new energy vehicles, the intelligent sensing module includes:

[0010] A voltage data acquisition unit, configured to collect real-time voltage data of the electrical equipment of the whole vehicle;

[0011] The current data acquisition unit is used to acquire the real-time current data of the vehicle's electrical equipment;

[0012] The temperature data acquisition unit is used to acquire the real-time working temperature of each electrical equipment;

[0013] The vibration data acquisition unit is used to acquire the real-time working vibration data of the motor corresponding to its electrical equipment;

[0014] The multi-dimensional data generation unit is used to integrate the real-time voltage data, real-time current data, real-time working temperature, and real-time working vibration data in the same time period based on the time axis to generate multi-dimensional electrical data.

[0015] Preferably, in a new energy vehicle electrical equipment detection system, the data processing module includes:

[0016] The data analysis unit is used to obtain the historical voltage data, historical current data, and historical working temperature data of different electrical equipment of the current model vehicle;

[0017] Calculate the average voltage and voltage standard deviation of each electrical equipment according to the historical current data;

[0018] Calculate the average current and current standard deviation of each electrical equipment according to the historical voltage data;

[0019] Calculate the average working temperature and temperature standard deviation of each electrical equipment according to the historical voltage data;

[0020] The first standardization unit is used to perform standardization processing on the real-time voltage data, real-time current data, and real-time working temperature respectively by using the Z-score algorithm based on the average voltage and voltage standard deviation, average current and current standard deviation, and average working temperature and temperature standard deviation to obtain the standardized voltage data, standardized current data, and standardized working temperature data;

[0021] The second standardization unit is used to obtain the factory parameters of the vehicle equipment, determine the corresponding working vibration intervals of the vehicle's electrical equipment respectively based on the factory parameters of the vehicle equipment, and perform normalization processing on the real-time working vibration data based on the upper and lower limits of the working vibration interval to obtain the standardized working vibration data;

[0022] The data integration unit is used to generate a standardized electrical data set based on the standardized voltage data, standardized current data, standardized working temperature data, and standardized working vibration data, and send it to the AI diagnosis module.

[0023] Preferably, in a new energy vehicle electrical equipment detection system, the AI diagnosis module includes:

[0024] A state evaluation unit, which is used to perform multi-dimensional analysis on standardized electrical data based on an AI fault prediction and diagnosis model, obtain analysis results in multiple dimensions, and comprehensively analyze the analysis results in multiple dimensions to determine the current health status of various electrical devices on the vehicle;

[0025] A state prediction unit, which is used to predict the short-term health status of electrical devices through an AI fault prediction and diagnosis model based on the current health status of each electrical device and the current electrical device operating data, and obtain device state prediction results;

[0026] A diagnostic report generation unit, which is used to generate a diagnostic report based on the current health status of electrical devices and the device state prediction results, send it to the user terminal and upload it to the cloud for storage;

[0027] A fault warning unit, which is used to send fault reminders for the corresponding electrical devices to the user and display and broadcast them on the vehicle control panel when there are faulty electrical devices or electrical devices with fault risks on the vehicle.

[0028] Preferably, in a new energy vehicle electrical equipment detection system, the AI diagnosis module further includes a model training unit for training the AI fault prediction and diagnosis model, including:

[0029] A data screening sub-unit, which is used to respectively obtain the corresponding historical data of various multi-dimensional electrical data and the maintenance data of the corresponding electrical devices, add labels to the historical data to obtain normal historical data and abnormal historical data;

[0030] A diagnostic special training sub-unit, which is used to classify the normal historical data and abnormal historical data based on the multi-dimensional electrical data types to generate training sets corresponding to various multi-dimensional electrical data;

[0031] Based on the training combination, train the preset fault diagnosis model respectively to obtain a preliminary training model, and classify the data in the training set corresponding to the electrical equipment types to obtain multiple electrical equipment special training sets;

[0032] Based on any electrical equipment special training data set, perform iterative training on the preliminary training model. After all the electrical equipment special training sets are trained, obtain the AI fault diagnosis model;

[0033] A prediction training sub-unit, which is used to train the preset fault prediction model based on the abnormal data linkage characteristics in the device abnormal prediction database to obtain the AI fault prediction model.

[0034] Preferably, in a new energy vehicle electrical equipment detection system, the state prediction unit includes:

[0035] A data feature extraction subunit, configured to respectively obtain the data fluctuation conditions of various multi-dimensional electrical data corresponding to each electrical device of the vehicle, align the various multi-dimensional electrical data based on the time axis, and determine the fluctuation relationship features between the various multi-dimensional electrical data according to the alignment result;

[0036] A fault prediction subunit, configured to analyze the fluctuation relationship features corresponding to each electrical device respectively based on an AI fault prediction model, and determine whether there is a fault risk for the electrical device, and obtain a fault risk prediction result of the electrical device;

[0037] Obtain the fault risk prediction results of all electrical devices of the vehicle and the multi-dimensional electrical data of the electrical devices with fault risks, and generate a device status prediction result.

[0038] Preferably, in a new energy vehicle electrical device detection system, the model training unit further includes:

[0039] A data processing subunit, configured to sort out the historical data of multiple electrical devices based on the time axis to respectively generate corresponding time series, and mark sequence abnormal nodes according to the labels carried by the historical data;

[0040] Centering on the sequence abnormal nodes, obtain the historical data of the sequence abnormal nodes of different electrical devices within a preset time range, generate a target time subsequence, and obtain multiple non-homogeneous data time subsequences based on the time region corresponding to the target time subsequence;

[0041] Generate fluctuation graphs respectively based on the target time subsequence and multiple non-homogeneous data time subsequences, and mark device abnormal points on the fluctuation graphs respectively according to the sequence abnormal nodes;

[0042] A data comparison subunit, configured to determine the data change features of different types of multi-dimensional electrical data before and after the device abnormal point marking based on the fluctuation graph;

[0043] Determine a first initial fluctuation point based on the data change features of the fluctuation graph corresponding to the target time subsequence;

[0044] Determine a second initial fluctuation point based on the data change features of the fluctuation graph corresponding to the non-homogeneous data time subsequence;

[0045] Vertically align and compare all the fluctuation graphs based on the time axis, and determine the time difference between the first initial fluctuation point and the second initial fluctuation point,

[0046] Extract the data fluctuation correlation relationship between different fluctuation graphs based on the time difference and in combination with the time point corresponding to the sequence abnormal node;

[0047] Obtain the data fluctuation correlation relationship of different electrical data of the same electrical equipment as the sequence abnormal nodes, and obtain the data fluctuation correlation relationship cluster;

[0048] Compare the data in the data fluctuation correlation relationship cluster and perform feature extraction to obtain the abnormal data linkage feature of the current electrical equipment;

[0049] Obtain the abnormal data linkage features corresponding to all electrical equipment of the vehicle to generate an equipment abnormal prediction database.

[0050] Preferably, in a new energy vehicle electrical equipment detection system, the remote interaction module includes:

[0051] A data access unit, which is used for maintenance personnel to input the dynamic secret key in the vehicle fault notification into the terminal device after receiving the vehicle fault notification, and access and obtain the diagnostic report;

[0052] A voice interaction unit, which is used for communication between maintenance personnel and users for fault handling;

[0053] A VR assistance unit, which is used for maintenance personnel to conduct visual remote guidance for vehicle maintenance through VR devices based on virtual reality technology.

[0054] Preferably, in a new energy vehicle electrical equipment detection system, the VR assistance unit includes:

[0055] A on-site data reproduction subunit, which is used to adjust the scenario environment of a preset vehicle three-dimensional model according to the real-time multi-dimensional electrical data of the vehicle and the on-site environment data to obtain a three-dimensional on-site maintenance scenario;

[0056] Real-time collect the on-site real-time maintenance data, and perform real-time adjustment on the vehicle three-dimensional model of the three-dimensional on-site maintenance scenario based on the real-time maintenance data;

[0057] A VR guidance subunit, which is used for maintenance personnel to conduct vehicle maintenance guidance according to the three-dimensional on-site maintenance scenario after wearing VR devices.

[0058] The present invention provides a new energy vehicle electrical equipment detection method, including:

[0059] Real-time collect the multi-dimensional electrical data of the vehicle;

[0060] Perform standardization processing on the multi-dimensional electrical data to generate a standardized electrical data set;

[0061] Analyze the standardized electrical data based on the AI fault prediction and diagnosis model, evaluate and predict the health status of various electrical equipment on the vehicle, and generate a diagnostic report;

[0062] Maintenance personnel access the diagnostic report based on the access of the terminal device and send fault handling guidance to the vehicle users.

[0063] Compared with the prior art, the present invention has at least the following beneficial effects:

[0064] The present invention collects multi-dimensional electrical data of the vehicle in real time, which can comprehensively reflect the real-time operation status of electrical equipment. Compared with the traditional single-parameter monitoring, it greatly improves the sensitivity to potential fault hazards of electrical equipment, ensures that even the slightest abnormalities can be captured in time, and lays a solid foundation for subsequent accurate diagnosis. Then, the multi-dimensional electrical data is standardized to unify the dimension and value range of different types of data, eliminate the analysis obstacles caused by data format and unit differences, enhance data compatibility, and provide a basis for the efficient processing of the subsequent AI diagnosis model. Subsequently, based on the AI fault prediction and diagnosis model, it can quickly perform in-depth analysis on the standardized electrical data. Compared with the traditional manual diagnosis or simple rule matching diagnosis methods, it greatly improves the diagnosis efficiency, can process a large amount of data in a short time, quickly and accurately evaluate the health status of various electrical equipment, predict potential faults, greatly shorten the fault troubleshooting time, improve the vehicle maintenance efficiency, provide strong support for preventive maintenance, arrange maintenance plans in advance, reduce the losses and inconveniences caused by equipment downtime. Finally, maintenance personnel can conveniently access and obtain the diagnostic report based on the terminal device. Without having to visit the vehicle site, they can quickly understand the detailed fault information of the vehicle's electrical equipment, ensure that users can understand the vehicle fault situation and corresponding handling measures in the first time, break the time and space limitations of vehicle maintenance. Especially in the case of emergency fault handling or when the vehicle is in a remote area, users can take effective fault repair methods according to the remote guidance of maintenance personnel in time, speed up the repair process, reduce the troubles and uneasiness brought to users by the fault, and effectively improve the satisfaction of users' vehicle use.

[0065] Other features and advantages of the present invention will be described in the subsequent description, and, in part, will be obvious from the description, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained by the structures specifically pointed out in the present application document.

[0066] The technical solutions of the present invention will be further described in detail below through the drawings and embodiments. Description of the Drawings

[0067] The drawings are used to provide a further understanding of the present invention, and constitute a part of the description. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0068] Figure 1 It is a structural diagram of an electrical equipment detection system for a new energy vehicle of the present invention;

[0069] Figure 2 This is the structural diagram of the intelligent sensing module of an electrical equipment detection system for a new energy vehicle according to the present invention;

[0070] Figure 3 This is the structural diagram of the data processing module of an electrical equipment detection system for a new energy vehicle according to the present invention;

[0071] Figure 4 This is the structural diagram of the AI diagnosis module of an electrical equipment detection system for a new energy vehicle according to the present invention;

[0072] Figure 5 This is the structural diagram of the remote interaction module of an electrical equipment detection system for a new energy vehicle according to the present invention. Specific embodiments

[0073] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for the purpose of illustrating and explaining the present invention, and are not intended to limit the present invention.

[0074] Embodiment 1:

[0075] The present invention provides an electrical equipment detection system for a new energy vehicle, as Figure 1 shown, including:

[0076] An intelligent sensing module for real-time collection of multi-dimensional electrical data of the vehicle;

[0077] A data processing module for standardizing multi-dimensional electrical data to generate a standardized electrical data set;

[0078] An AI diagnosis module for analyzing the standardized electrical data based on an AI fault prediction and diagnosis model to evaluate and predict the health status of various electrical equipment on the vehicle, and generate a diagnosis report;

[0079] A remote interaction module for maintenance personnel to access the diagnosis report based on a terminal device and send fault handling guidance to vehicle users.

[0080] Advantages of the above technical solution: The present invention collects multi-dimensional electrical data of the vehicle in real time, which can comprehensively reflect the real-time operating status of electrical equipment. Compared with traditional single-parameter monitoring, it greatly improves the sensitivity to potential faults and hidden dangers of electrical equipment, ensuring that even the slightest abnormalities can be captured in time, laying a solid foundation for subsequent accurate diagnosis. Then, the multi-dimensional electrical data is standardized to unify the dimension and value range of different types of data, eliminating the analysis obstacles caused by data format and unit differences, enhancing data compatibility, and providing a basis for the efficient processing of the subsequent AI diagnosis model. Subsequently, based on the AI fault prediction and diagnosis model, it can quickly perform in-depth analysis on the standardized electrical data. Compared with traditional manual diagnosis or simple rule-based matching diagnosis methods, it greatly improves the diagnosis efficiency, can process a large amount of data in a short time, quickly and accurately evaluate the health status of various electrical equipment, and predict potential faults, significantly shortening the fault troubleshooting time, improving the vehicle maintenance efficiency, providing strong support for preventive maintenance, arranging maintenance plans in advance, and reducing the losses and inconveniences caused by equipment downtime. Finally, maintenance personnel can conveniently access and obtain the diagnosis report based on the terminal device, and can quickly understand the detailed fault information of the vehicle's electrical equipment without having to go to the vehicle site, ensuring that users can understand the vehicle fault situation and corresponding treatment measures in the first time, breaking the time and space limitations of vehicle maintenance. Especially in the case of emergency fault handling or when the vehicle is in a remote area, users can take effective fault repair methods according to the remote guidance of maintenance personnel in a timely manner, accelerating the repair process, reducing the troubles and uneasiness brought to users by the fault, and effectively improving the satisfaction of users' vehicle use.

[0081] Embodiment 2:

[0082] Based on Embodiment 1, the intelligent sensing module, such as Figure 2 shown, includes:

[0083] A voltage data acquisition unit for collecting the real-time voltage data of the vehicle's electrical equipment;

[0084] A current data acquisition unit for collecting the real-time current data of the vehicle's electrical equipment;

[0085] A temperature data acquisition unit for collecting the real-time working temperature of each electrical equipment;

[0086] A vibration data acquisition unit for collecting the real-time working vibration data of the motors corresponding to the electrical equipment;

[0087] A multi-dimensional data generation unit for integrating and generating multi-dimensional electrical data based on the time axis for the real-time voltage data, real-time current data, real-time working temperature, and real-time working vibration data in the same period.

[0088] Advantages of the above technical solution: Through the voltage data acquisition unit, current data acquisition unit, temperature data acquisition unit, and vibration data acquisition unit, the present invention realizes the synchronous acquisition of multi-dimensional electrical data for comprehensive monitoring of vehicle electrical equipment. Finally, through the multi-dimensional data generation unit, the collected voltage, power, temperature, and vibration data are integrated to generate multi-dimensional electrical data, achieving the organic integration of different data, greatly improving the sensitivity to potential faults and hidden dangers of electrical equipment, ensuring that even the slightest anomalies can be captured in a timely manner, and laying a solid foundation for subsequent accurate diagnosis.

[0089] Embodiment 3:

[0090] Based on Embodiment 1, the data processing module, as Figure 3 shown, includes:

[0091] The data analysis unit is used to obtain the historical voltage data, historical current data, and historical operating temperature data of different electrical equipment of the current model of vehicle;

[0092] Calculate the average voltage and voltage standard deviation of each electrical equipment respectively according to the historical current data;

[0093] Calculate the average current and current standard deviation of each electrical equipment respectively according to the historical voltage data;

[0094] Calculate the average operating temperature and temperature standard deviation of each electrical equipment respectively according to the historical voltage data;

[0095] The first standardization unit is used to perform standardization processing on the real-time voltage data, real-time current data, and real-time operating temperature respectively by using the Z-score algorithm based on the average voltage and voltage standard deviation, average current and current standard deviation, and average operating temperature and temperature standard deviation, to obtain the standardized voltage data, standardized current data, and standardized operating temperature data;

[0096] The second standardization unit is used to obtain the factory parameters of the vehicle equipment, and based on the factory parameters of the vehicle equipment, determine the corresponding operating vibration intervals of the vehicle's electrical equipment respectively, and perform normalization processing on the real-time operating vibration data based on the upper and lower limits of the operating vibration intervals to obtain the standardized operating vibration data;

[0097] The data integration unit is used to generate a standardized electrical data set based on the standardized voltage data, standardized current data, standardized operating temperature data, and standardized operating vibration data, and send it to the AI diagnosis module.

[0098] Advantages of the above technical solution: The present invention standardizes multi-dimensional electrical data through different standardization methods. The Z-score algorithm is used to standardize real-time voltage, current, and working temperature data based on the average voltage, voltage standard deviation, average current, current standard deviation, average working temperature, and temperature standard deviation, eliminating the differences in dimension and scale between different data types. This facilitates the comparison and analysis of data from different electrical devices under the same standard, improving the comparability and usability of the data. For example, the unit of voltage data may be volts, while the unit of temperature data is degrees Celsius. Through standardization, they can be operated within the same numerical range, providing convenience for subsequent data analysis and model training. At the same time, the factory parameters of vehicle equipment are obtained, and based on this, the working vibration range corresponding to the vehicle's electrical equipment is determined. The real-time working vibration data is normalized. By processing the vibration data based on the factory standard of the electrical equipment, the vibration data can be closely combined with the design requirements of the equipment, ensuring that the processed vibration data can accurately reflect the deviation of the actual operating state of the equipment from the design standard. For example, if the factory design of a certain motor requires its vibration amplitude to be within a certain range, through normalization, it can be intuitively judged whether the current vibration of the motor exceeds the normal range, providing a more accurate basis for fault diagnosis. Standardizing multi-dimensional electrical data separately can enhance the generalization ability of the model for data under different working conditions and environments, enabling it to more accurately identify the fault states of equipment. While enhancing data compatibility, it also improves the accuracy of data standardization, providing more reliable data support for comprehensively evaluating the health status of equipment in the future.

[0099] Embodiment 4:

[0100] Based on Embodiment 1, the AI diagnosis module, as Figure 4 shown, includes:

[0101] A state evaluation unit for performing multi-dimensional analysis on the standardized electrical data based on the AI fault prediction and diagnosis model, obtaining analysis results in multiple dimensions, and comprehensively analyzing the analysis results in multiple dimensions to determine the current health status of various electrical devices on the vehicle;

[0102] A state prediction unit for predicting the short-term health status of the electrical device based on the current health status of each electrical device and the current working data of the electrical device through the AI fault prediction and diagnosis model, obtaining the device state prediction result;

[0103] A diagnosis report generation unit for generating a diagnosis report based on the current health status of the electrical device and the device state prediction result, sending it to the user terminal and uploading it to the cloud for storage;

[0104] A fault warning unit is used to send a fault reminder for the corresponding electrical equipment to the user and display and broadcast it on the vehicle control panel when there is a faulty electrical equipment or a faulty-risk electrical equipment in the vehicle.

[0105] Advantages of the above technical solution: Based on the AI fault prediction and diagnosis model, the present invention can quickly perform in-depth analysis on standardized electrical data. Compared with traditional manual diagnosis or simple rule-based matching diagnosis methods, it greatly improves the diagnosis efficiency, can process a large amount of data in a short time, quickly and accurately evaluate the health status of various electrical equipment, predict potential faults, significantly shorten the fault troubleshooting time, improve the vehicle maintenance efficiency, provide strong support for preventive maintenance, arrange maintenance plans in advance, reduce the losses caused by equipment downtime, and generate detailed diagnostic reports and send them to the user terminal and the cloud, enabling users and maintenance personnel to clearly understand the current health status and future prediction of the vehicle's electrical equipment. When a faulty electrical equipment or a faulty-risk electrical equipment is detected in the vehicle, a fault reminder is immediately sent to the user and displayed and broadcast on the vehicle control panel, intuitively and timely reminding the user of the vehicle fault, which helps the user quickly locate the faulty equipment, enables the user to be aware of the vehicle's abnormal situation in the first time, and take corresponding safety measures, such as slowing down and finding a safe place to park as soon as possible, effectively avoiding more serious consequences caused by not detecting the fault in time, and ensuring the personal safety of the user and the safe operation of the vehicle.

[0106] Embodiment 5:

[0107] Based on Embodiment 4, the AI diagnosis module, as Figure 4 shown, further includes a model training unit for training the AI fault prediction and diagnosis model, including:

[0108] A data screening sub-unit for respectively obtaining the corresponding historical data of various multi-dimensional electrical data and the maintenance data of the corresponding electrical equipment, adding labels to the historical data to obtain normal historical data and abnormal historical data;

[0109] A diagnosis special training sub-unit for classifying the normal historical data and abnormal historical data based on the multi-dimensional electrical data types to generate training sets corresponding to various multi-dimensional electrical data;

[0110] Based on the training combination, training the preset fault diagnosis model respectively to obtain a preliminary training model, and classifying based on the data in the training set corresponding to the electrical equipment types to obtain multiple electrical equipment special training sets;

[0111] Based on any electrical equipment special training data set, performing iterative training on the preliminary training model, and after all electrical equipment special training sets are trained, obtaining the AI fault diagnosis model;

[0112] A prediction training subunit, configured to train a preset fault prediction model based on abnormal data linkage features in a device anomaly prediction database to obtain an AI fault prediction model.

[0113] Beneficial effects of the above technical solution: The present invention obtains historical data of multi-dimensional electrical data and maintenance data of corresponding electrical devices, and adds labels to the historical data to accurately distinguish normal historical data from abnormal historical data. This provides clear and accurate labeled data for subsequent model training, enabling the model to learn in a supervised environment and effectively improving the quality of training data. High-quality training data is the basis for constructing a high-performance AI fault diagnosis model, which can help the model better learn the feature differences between the normal and abnormal states of electrical devices, thereby improving the diagnostic accuracy and reliability of the model; Classify normal and abnormal historical data based on multi-dimensional electrical data types to generate training sets corresponding to various multi-dimensional electrical data, enabling the model to learn according to the data characteristics of different types and better capture the fault information contained in each data dimension. For example, for the training set of voltage data, the model can focus on learning voltage-related fault features, such as the association between abnormal situations such as voltage sudden changes, overvoltage or undervoltage and device faults, thereby improving the diagnostic ability for voltage-related faults and enhancing the pertinence of diagnosis. Subsequently, use multiple special training sets for electrical devices to iteratively train the preliminary training model, enabling the model to be optimized according to the characteristics of different electrical devices and improving the accuracy and adaptability of fault diagnosis for various electrical devices. For example, for a battery management system and a motor controller, their fault manifestations and related data characteristics are different, and special training can enable the model to master these differences respectively to achieve accurate diagnosis of different devices. Training a preset fault prediction model based on abnormal data linkage features in a device anomaly prediction database enables the model to learn the correlation change trend between various data before a device fault occurs, be able to predict the occurrence of a device fault in advance, provide a more timely fault warning for users, and greatly improve the safety and reliability of vehicle use. The present invention divides the AI fault prediction and diagnosis model into two sub-models (AI fault diagnosis model and AI fault prediction model) for training respectively, which not only reduces the complexity of the model and the risk of overfitting, but also facilitates the targeted training of the model, further improving the reliability of the model.

[0114] Embodiment 6:

[0115] On the basis of Embodiment 4, the state prediction unit includes:

[0116] A data feature extraction subunit, configured to respectively obtain the data fluctuation conditions of various multi-dimensional electrical data corresponding to each electrical device of the vehicle, align the various multi-dimensional electrical data based on the time axis, and determine the fluctuation relationship features between the various multi-dimensional electrical data according to the alignment result;

[0117] A fault prediction sub - unit, which is used to analyze the fluctuation relationship characteristics corresponding to each electrical device based on the AI fault prediction model, and judge whether there is a fault risk in the electrical device, so as to obtain the fault risk prediction result of the electrical device;

[0118] Obtain the fault risk prediction results of all electrical devices in the vehicle and the multi - dimensional electrical data of the electrical devices with fault risks, and generate the equipment status prediction results.

[0119] Beneficial effects of the above technical solution: By obtaining the data fluctuation conditions of various multi - dimensional electrical data corresponding to each electrical device in the vehicle and aligning the multi - dimensional electrical data based on the time axis, the present invention can accurately determine the fluctuation relationship characteristics between different types of multi - dimensional electrical data. For example, during the operation of the motor, the voltage fluctuation may cause corresponding changes in current and temperature. Through this alignment and analysis, the specific association patterns between them can be discovered, which helps to reveal potential fault causes and early fault signals, can provide a more comprehensive and accurate basis for equipment health assessment, provide more powerful support for subsequent fault prediction and diagnosis, and then analyze the association characteristics between multi - dimensional electrical data through the AI fault prediction model to predict potential vehicle faults, greatly shortening the fault troubleshooting time, improving the vehicle maintenance efficiency, providing strong support for preventive maintenance, arranging maintenance plans in advance, and reducing the losses and inconveniences caused by equipment downtime.

[0120] Embodiment 7:

[0121] On the basis of Embodiment 5, the model training unit further includes:

[0122] A data processing sub - unit, which is used to sort the historical data of multiple electrical devices based on the time axis to generate corresponding time series respectively, and mark the sequence abnormal nodes according to the labels carried by the historical data;

[0123] Taking the sequence abnormal node as the center, obtain the historical data of the sequence abnormal nodes of different electrical devices within a preset time range to generate a target time subsequence, and obtain multiple non - homogeneous data time subsequences based on the time region corresponding to the target time subsequence;

[0124] Generate fluctuation graphs based on the target time subsequence and multiple non - homogeneous data time subsequences respectively, and mark the device abnormal points on the fluctuation graphs according to the sequence abnormal nodes;

[0125] A data comparison sub - unit, which is used to determine the data change characteristics of different types of multi - dimensional electrical data before and after the device abnormal point marking based on the fluctuation graphs;

[0126] Determine the first initial fluctuation point based on the data change characteristics of the fluctuation graph corresponding to the target time subsequence;

[0127] Determine the second initial fluctuation point based on the data change characteristics of the fluctuation graph corresponding to the non - homogeneous data time subsequence;

[0128] Based on the time axis, perform vertical alignment and comparison on all the fluctuation graphs to determine the time difference between the first initial fluctuation point and the second initial fluctuation point.

[0129] Based on the time difference, combined with the time points corresponding to the sequence anomaly nodes, extract the data fluctuation correlation relationship between different fluctuation graphs;

[0130] Obtain the data fluctuation correlation relationships of different electrical data of the same electrical equipment as the sequence anomaly nodes to get a data fluctuation correlation relationship cluster;

[0131] Compare the data within the data fluctuation correlation relationship cluster and perform feature extraction to obtain the abnormal data linkage feature of the current electrical equipment;

[0132] Obtain the abnormal data linkage features corresponding to all the electrical equipment of the vehicle to generate an equipment anomaly prediction database.

[0133] Among them, the sequence anomaly node refers to the time point when the equipment fails.

[0134] In this embodiment, the first initial fluctuation point and the second initial fluctuation point represent the starting times when abnormal fluctuations begin to appear in different equipment or different types of data before the occurrence of a fault.

[0135] Among them, the first initial fluctuation point refers to the starting time when abnormal fluctuations begin to appear in the abnormal historical data of the corresponding type of the target time subsequence before the occurrence of a fault;

[0136] The second initial fluctuation point refers to the starting time when abnormal fluctuations begin to appear in the abnormal historical data of the corresponding type of the non - homogeneous data time subsequence before the occurrence of a fault.

[0137] In this embodiment, the non - homogeneous data time subsequence refers to the time sequence corresponding to other data of the same time period with different electrical data types from the abnormal historical data corresponding to the target time subsequence. For example, when the target time subsequence is the current data of an electrical equipment, the non - homogeneous data time subsequences are the voltage, temperature, and vibration data of the electrical equipment in the same time period as the target time subsequence, and these three types of data will respectively generate different non - homogeneous data time subsequences; when the target time subsequence is temperature data, the non - homogeneous data time subsequences are the current, voltage, and vibration data in the same time period as the target time subsequence.

[0138] Beneficial effects of the above technical solution: Based on the time axis, the present invention organizes the historical data of multiple electrical devices into corresponding time series, making the originally chaotic historical data orderly and having coherence in the time dimension, which helps to clearly show the changing trend of the operating state of electrical devices at different times, and marks the sequence abnormal nodes according to the tags carried by the historical data to accurately locate the key time points when the device fails or is abnormal, providing a clear direction for subsequent targeted analysis, greatly improving the efficiency of data processing and analysis. Then, taking the sequence abnormal nodes as the center, historical data within a preset time range is obtained to generate target time subsequences, and further multiple non-homogeneous data time subsequences are obtained, focusing the abnormal analysis of the device on the core period when the device abnormality occurs, comprehensively covering various types of data related to the abnormality. For example, when a motor fails, not only the changes in various electrical data (such as voltage, current, temperature, vibration) of the motor itself before and after the failure are obtained, but also the data of other associated devices (such as controllers, sensors) within the same time period are obtained, realizing multi-dimensional and cross-device data collection, and being able to comprehensively and meticulously depict the data dynamics of the entire electrical system when the abnormality occurs, providing a basis for the extraction of multi-dimensional data correlation features. Then, the target time subsequences and non-homogeneous data time subsequences are respectively generated into fluctuation diagrams, and the device abnormal points are marked on the fluctuation diagrams, converting complex data into intuitive and easy-to-understand visual graphs, which can clearly show the ups and downs of the data in the time dimension, and the marking of the device abnormal points highlights the specific moment and data characteristics when the failure occurs;Subsequently, based on the fluctuation graph, the data change characteristics of different types of multi-dimensional electrical data before and after the device anomaly point marking are determined, which facilitates determining the specific change conditions of various data of the electrical equipment when a fault occurs. Then, based on the data change characteristics of the fluctuation graphs corresponding to the target time subsequence and the non-homogeneous data time subsequence respectively, the first initial fluctuation point and the second initial fluctuation point are determined, laying a foundation for capturing the early signs of equipment faults in advance. Finally, based on the time axis, all the fluctuation graphs are vertically aligned and compared to determine the time difference between the first initial fluctuation point and the second initial fluctuation point, and combined with the time points corresponding to the sequence anomaly nodes, the data fluctuation correlation relationship between different fluctuation graphs is extracted, realizing the cross-data type data comparison and analysis of the same equipment, which can reveal the internal connections between electrical equipment and between different types of data during the fault occurrence process, obtain the data fluctuation correlation relationship (for example, it is found that there is a specific time difference between the initial fluctuation of the motor current and the initial fluctuation of the controller output voltage, and this time difference is consistent in multiple similar fault events), obtain the data fluctuation correlation relationship of different electrical data of the same electrical equipment as the sequence anomaly nodes, form a data fluctuation correlation relationship cluster, compare and extract the characteristics of the data fluctuation correlation relationship within the cluster to obtain the abnormal data linkage characteristics of the current electrical equipment, complete the abnormal data linkage characteristics of all fault manifestations of the electrical equipment, integrate the abnormal data linkage characteristics corresponding to all the electrical equipment of the vehicle to generate an equipment anomaly prediction database, which provides rich and comprehensive training data for the training of the AI fault prediction model. By learning the data fluctuation correlation characteristics between multi-dimensional electrical data, the model can more accurately predict the occurrence of equipment faults, improve the accuracy and generalization ability of fault prediction, and provide a strong guarantee for the reliability and safety of the vehicle electrical system.

[0139] Embodiment 8:

[0140] Based on Embodiment 1, the remote interaction module, as Figure 5 shown, includes:

[0141] A data access unit, which is used for maintenance personnel to input the dynamic secret key in the vehicle fault notification into the terminal device after receiving the vehicle fault notification, and access and obtain the diagnostic report;

[0142] A voice interaction unit, which is used for communication between maintenance personnel and users for fault handling;

[0143] A VR assistance unit, which is used for the maintenance personnel to conduct visual remote guidance on vehicle maintenance through VR devices based on virtual reality technology.

[0144] Beneficial effects of the above technical solution: After receiving a vehicle fault notification, the maintenance personnel of the present invention can quickly access and obtain the diagnostic report by inputting the dynamic secret key into the terminal device, greatly shortening the time to obtain vehicle fault information. This enables the maintenance personnel to master the detailed fault conditions of the vehicle's electrical equipment in a timely manner without having to be on-site. At the same time, the diagnostic report is accessed using the dynamic secret key method, providing strong protection for data security, protecting the privacy of vehicle users and the security of vehicle information, and ensuring the confidentiality and integrity of data during transmission and storage. The maintenance personnel and the user can communicate in real time about fault handling through the voice interaction unit, enabling the maintenance personnel to quickly understand the specific situation when the fault occurred, such as the driving state of the vehicle, the details before and after the fault occurred, etc. At the same time, the maintenance personnel can clearly explain the fault cause and handling steps to the user, and the user can also give timely feedback. The efficient communication between the two parties helps to accurately judge the fault and formulate a suitable maintenance plan, avoiding maintenance delays or misjudgments caused by poor information transmission. And based on virtual reality technology, the maintenance personnel can conduct visual remote guidance for vehicle maintenance through VR devices, breaking the spatial limit. The maintenance personnel can observe the vehicle fault location as if they were on-site and use the interaction function of the VR device to mark the fault points in real time and demonstrate the maintenance steps for on-site operators (such as users or on-site maintenance assistants).

[0145] Embodiment 9:

[0146] Based on Embodiment 8, the VR assistance unit includes:

[0147] A on-site data reproduction sub-unit, configured to adjust the preset three-dimensional vehicle model according to the real-time multi-dimensional electrical data of the vehicle and the on-site environmental data to obtain a three-dimensional on-site maintenance scenario;

[0148] Real-time collect the on-site real-time maintenance data, and perform real-time adjustment on the three-dimensional vehicle model of the three-dimensional on-site maintenance scenario based on the real-time maintenance data;

[0149] A VR guidance sub-unit, configured to, after the maintenance personnel wear the VR device and enter the three-dimensional on-site maintenance scenario, conduct vehicle maintenance guidance according to the three-dimensional on-site maintenance scenario.

[0150] Beneficial effects of the above technical solution: According to the real-time multi-dimensional electrical data of the vehicle and the on-site environmental data, the present invention adjusts the preset three-dimensional vehicle model for the scenario environment to generate an extremely realistic three-dimensional scenario of the maintenance site, making the maintenance personnel seem to be in the real maintenance site, greatly enhancing the immersion and realism of remote maintenance guidance, improving the accuracy of fault diagnosis. Subsequently, the on-site real-time maintenance data is collected in real time, and based on this, the three-dimensional vehicle model of the three-dimensional scenario of the maintenance site is adjusted in real time to ensure that the maintenance personnel always grasp the latest maintenance progress, ensure the coherence and accuracy of the maintenance process, avoid operation errors or maintenance delays caused by information lag, and help the maintenance personnel optimize the maintenance plan according to the actual maintenance situation. The VR guidance sub-unit breaks through the space limit and realizes the efficient remote collaboration between the maintenance personnel and the on-site operators.

[0151] Embodiment 10:

[0152] The present invention provides a method for detecting electrical equipment of a new energy vehicle, including:

[0153] Collecting multi-dimensional electrical data of the vehicle in real time;

[0154] Performing standardization processing on the multi-dimensional electrical data to generate a standardized electrical data set;

[0155] Analyzing the standardized electrical data based on the AI fault prediction and diagnosis model to evaluate and predict the health status of various electrical equipment on the vehicle, and generating a diagnostic report;

[0156] The maintenance personnel access the diagnostic report based on the terminal device and send fault handling guidance to the vehicle user.

[0157] Advantages of the above technical solution: The present invention collects multi-dimensional electrical data of the vehicle in real time, which can comprehensively reflect the real-time operating status of electrical equipment. Compared with traditional single-parameter monitoring, it greatly improves the sensitivity to potential faults and hidden dangers of electrical equipment, ensuring that even the slightest abnormalities can be captured in time, laying a solid foundation for subsequent accurate diagnosis. Then, the multi-dimensional electrical data is standardized to unify the dimension and value range of different types of data, eliminating the analysis obstacles caused by data format and unit differences, enhancing data compatibility, and providing a basis for the efficient processing of the subsequent AI diagnosis model. Subsequently, based on the AI fault prediction and diagnosis model, it can quickly conduct in-depth analysis of the standardized electrical data. Compared with traditional manual diagnosis or simple rule-based matching diagnosis methods, it greatly improves the diagnosis efficiency, can process a large amount of data in a short time, quickly and accurately evaluate the health status of various electrical equipment, and predict potential faults, significantly shortening the fault troubleshooting time, improving the vehicle maintenance efficiency, providing strong support for preventive maintenance, arranging maintenance plans in advance, and reducing the losses and inconveniences caused by equipment downtime. Finally, maintenance personnel can conveniently access and obtain the diagnostic report based on the terminal device, quickly understand the detailed fault information of the vehicle's electrical equipment without having to go to the vehicle site, ensuring that users can understand the vehicle fault situation and corresponding treatment measures in the first time, breaking the time and space limitations of vehicle maintenance. Especially in the case of emergency fault handling or when the vehicle is in a remote area, users can take effective fault repair methods according to the remote guidance of maintenance personnel in time, accelerating the repair process, reducing the troubles and uneasiness brought to users by faults, and effectively improving the satisfaction of users' vehicle use.

[0158] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A new energy vehicle electrical equipment detection system, characterized in that: include: Intelligent sensor module, used to collect multi-dimensional electrical data of the car in real time; A data processing module is used to perform standardized processing on multi-dimensional electrical data to generate a standardized electrical data set; AI diagnostic module, which is used to analyze standardized electrical data based on AI fault prediction and diagnosis models, determine the health status of various electrical equipment on the car, evaluate and predict it, and generate diagnostic reports; The remote interaction module is used by maintenance personnel to obtain diagnostic reports based on terminal devices and send fault handling instructions to car users.

2. A new energy vehicle electrical equipment detection system according to claim 1, characterized in that: Smart sensor module, including: Voltage data acquisition unit, used to collect real-time voltage data of the vehicle's electrical equipment; Current data acquisition unit, used to collect real-time current data of the vehicle's electrical equipment; Temperature data acquisition unit, used to collect the real-time operating temperature of each electrical device; A vibration data acquisition unit, used to collect real-time working vibration data of the motor corresponding to the electrical equipment; The multi-dimensional data generation unit is used to integrate the real-time voltage data, real-time current data, real-time working temperature and real-time working vibration data of the same time period based on the time axis to generate multi-dimensional electrical data.

3. A new energy vehicle electrical equipment detection system according to claim 1, characterized in that: Data processing module, including: A data analysis unit, used to obtain historical voltage data, historical current data, and historical operating temperature data of different electrical devices of the current model of automobile; Calculate the average voltage and voltage standard deviation of each electrical device according to the historical current data; Calculate the average current and current standard deviation of each electrical device according to the historical voltage data; Calculate the average operating temperature and temperature standard deviation of each electrical device according to the historical voltage data; A first standardization unit is used to standardize the real-time voltage data, the real-time current data, and the real-time operating temperature by using a Z-score algorithm based on the average voltage and the voltage standard deviation, the average current and the current standard deviation, and the average operating temperature and the temperature standard deviation, to obtain standardized voltage data, standardized current data, and standardized operating temperature data; A second standardization unit is used to obtain factory parameters of vehicle equipment, determine working vibration ranges corresponding to the electrical equipment of the whole vehicle based on the factory parameters of the vehicle equipment, and perform normalization processing on the real-time working vibration data based on upper and lower limits of the working vibration range to obtain standardized working vibration data; The data integration unit is used to generate a standardized electrical data set based on the standardized voltage data, the standardized current data, the standardized operating temperature data and the standardized operating vibration data, and send the data set to the AI ​​diagnosis module.

4. A new energy vehicle electrical equipment detection system according to claim 1, characterized in that: AI diagnostic module, including: The status assessment unit is used to perform multi-dimensional analysis on standardized electrical data based on the AI ​​fault prediction and diagnosis model, obtain analysis results in multiple dimensions, and conduct comprehensive analysis on the analysis results in multiple dimensions to determine the current health status of various electrical devices on the vehicle; The state prediction unit is used to predict the short-term health state of electrical equipment based on the current health state of each electrical equipment and the current working data of the electrical equipment through the AI ​​fault prediction and diagnosis model to obtain the equipment state prediction result; A diagnostic report generation unit, used to generate a diagnostic report based on the current health status of the electrical equipment and the equipment status prediction result, send it to the user terminal and upload it to the cloud for storage; The fault warning unit is used to send fault reminders of the corresponding electrical equipment to the user and display them on the car control panel when there are faulty electrical equipment or electrical equipment with a risk of failure in the car.

5. A new energy vehicle electrical equipment detection system according to claim 4, characterized in that: The AI ​​diagnosis module also includes a model training unit for training the AI ​​fault prediction and diagnosis model, including: The data identification subunit is used to obtain the corresponding historical data of various multi-dimensional electrical data and the maintenance data of the corresponding electrical equipment, add labels to the historical data, and obtain normal historical data and abnormal historical data; A diagnostic special training subunit is used to classify normal historical data and abnormal historical data based on multi-dimensional electrical data types, and generate training sets corresponding to various multi-dimensional electrical data; Based on the training combination, the preset fault diagnosis models are trained respectively to obtain a preliminary training model, and the data in the training set are classified according to the type of electrical equipment to obtain multiple electrical equipment special training sets; Based on any electrical equipment special training data set, the preliminary training model is iteratively trained. After the training of all electrical equipment special training sets is completed, the AI ​​fault diagnosis model is obtained; The prediction training subunit is used to train the preset fault prediction model based on the abnormal data linkage characteristics in the equipment abnormality prediction database to obtain an AI fault prediction model.

6. A new energy vehicle electrical equipment detection system according to claim 4, characterized in that: State prediction unit, including: The data feature extraction subunit is used to respectively obtain data fluctuations of various multi-dimensional electrical data corresponding to various electrical devices of the vehicle, align the various multi-dimensional electrical data based on the time axis, and determine the fluctuation relationship characteristics between the various multi-dimensional electrical data according to the alignment results; The fault prediction subunit is used to analyze the fluctuation relationship characteristics corresponding to each electrical equipment based on the AI ​​fault prediction model, and determine whether the electrical equipment has a fault risk, and obtain the fault risk prediction result of the electrical equipment; Obtain the failure risk prediction results of all electrical equipment in the vehicle and the multi-dimensional electrical data of electrical equipment with failure risks, and generate equipment status prediction results.

7. A new energy vehicle electrical equipment detection system according to claim 5, characterized in that: The model training unit also includes: The data processing subunit is used to sort out the historical data of multiple electrical devices based on the time axis to generate corresponding time series, and mark abnormal nodes in the sequence according to the labels carried by the historical data; Taking the sequence abnormal node as the center, the historical data of the sequence abnormal nodes of different electrical equipment within a preset time range are obtained to generate a target time subsequence, and based on the time region corresponding to the target time subsequence, multiple non-similar data time subsequences are obtained; Generate fluctuation graphs based on the target time subsequence and multiple non-similar data time subsequences, and mark equipment abnormal points on the fluctuation graphs according to abnormal nodes in the sequences; A data comparison subunit is used to determine the data change characteristics of different types of multi-dimensional electrical data before and after the abnormal point of the equipment is marked based on the fluctuation diagram; Determine a first initial fluctuation point based on data change characteristics of the fluctuation graph corresponding to the target time subsequence; Determining a second initial fluctuation point based on data change characteristics of the fluctuation graph corresponding to the non-similar data time subsequences; Based on the time axis, all fluctuation graphs are aligned and compared vertically to determine the time difference between the first initial fluctuation point and the second initial fluctuation point. Based on the time difference, combined with the time points corresponding to the abnormal nodes of the sequence, the data fluctuation correlation relationship between different fluctuation graphs is extracted; Obtain data fluctuation correlation relationships of different electrical data of the same electrical equipment as sequence abnormal nodes to obtain data fluctuation correlation relationship clusters; Compare the data in the data fluctuation correlation cluster and extract features to obtain the abnormal data linkage features of the current electrical equipment; The abnormal data linkage features corresponding to all the electrical equipment of the car are obtained to generate an equipment abnormality prediction database.

8. A new energy vehicle electrical equipment detection system according to claim 1, characterized in that: Remote interaction module, including: A data access unit is used for maintenance personnel to input the dynamic key in the vehicle fault notification into the terminal device after receiving the vehicle fault notification, and access and obtain the diagnostic report; Voice interaction unit, used for troubleshooting communication between maintenance personnel and users; The VR auxiliary unit is used to provide visual remote guidance of automobile maintenance to maintenance personnel through VR equipment based on virtual reality technology.

9. A new energy vehicle electrical equipment detection system according to claim 8, characterized in that: VR auxiliary unit, including: The on-site data reproduction subunit is used to adjust the scene environment of the preset three-dimensional vehicle model according to the real-time multi-dimensional electrical data of the vehicle and the on-site environmental data to obtain the three-dimensional scene of the maintenance site; Collect real-time maintenance data on site in real time, and adjust the three-dimensional model of the car in the three-dimensional scene of the maintenance site in real time based on the real-time maintenance data; The VR guidance subunit is used for maintenance personnel to provide vehicle maintenance guidance according to the three-dimensional scene of the maintenance site after wearing VR equipment.

10. A method for detecting electrical equipment of new energy vehicles, characterized in that: include: Collect multi-dimensional electrical data of the car in real time; Standardize multi-dimensional electrical data to generate standardized electrical data sets; Analyze standardized electrical data based on AI fault prediction and diagnosis models, determine the health status of various electrical equipment on the car, evaluate and predict it, and generate a diagnostic report; Maintenance personnel access the diagnostic report based on the terminal device and send fault handling instructions to the car user.

Images