Battery insulation abnormity fault detection method and device, program product and medium

Abstract

The invention discloses a battery insulation abnormity fault detection method and device, a program product and a medium. The method comprises the following steps: acquiring battery national standard data of a plurality of to-be-detected vehicles; inputting the battery national standard data of the plurality of to-be-detected vehicles into a fault detection model to obtain an early warning result for a battery insulation fault of each to-be-detected vehicle, the fault detection model including a pre-trained random forest model and a logistic regression model; screening the early warning result again to finally determine a fault vehicle with a battery insulation fault in the plurality of to-be-detected vehicles; and determining the fault type of each fault vehicle according to the charging state of each fault vehicle and the operation state of the voltage converter. According to the scheme, the early warning precision of the battery insulation abnormity fault is high, the generalization ability is high, the real-time performance is high, and the scheme is suitable for battery fault early warning in the fields of electric automobiles, energy storage systems and the like and is high in universality.

Description

Technical Field

[0001] The present application belongs to the field of battery detection, and in particular, relates to a battery insulation abnormality fault detection method, device, program product and medium. Background Art

[0002] With the widespread application of electric vehicles and energy storage systems, the safety and stability of batteries have become increasingly prominent. Abnormal battery insulation failure is one of the important reasons for battery failure, which may lead to a decrease in battery performance and even cause safety accidents. Therefore, early warning of abnormal battery insulation failure is of great significance to ensure the normal operation of the battery system and prevent safety accidents.

[0003] In the prior art, most of the warning methods for abnormal battery insulation faults are based on traditional physical models or empirical rules. However, this method has problems such as low accuracy and poor generalization ability for warning abnormal battery insulation faults. Summary of the invention

[0004] The embodiments of the present application provide a battery insulation abnormality fault detection method, device, program product and medium, thereby improving the accuracy of battery insulation abnormality fault detection at least to a certain extent.

[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0006] According to a first aspect of an embodiment of the present application, a battery insulation abnormal failure method is provided, the method comprising: obtaining national standard battery data of multiple vehicles to be detected; inputting the national standard battery data of multiple vehicles to be detected into a fault detection model to obtain a warning result of battery insulation failure for each vehicle to be detected, wherein the fault detection model comprises a pre-trained random forest model and a logistic regression model; re-screening the warning result to ultimately determine the faulty vehicle with battery insulation failure among the multiple vehicles to be detected; determining the fault type of each faulty vehicle according to the charging status of each faulty vehicle and the operating status of the voltage converter.

[0007] In some embodiments of the present application, based on the aforementioned scheme, the method also includes: the method also includes a training step for a random forest model, and the training step of the random forest model includes: collecting historical battery national standard data of multiple vehicles during historical operation; after preprocessing the historical battery national standard data, constructing a training set based on the preprocessed historical battery national standard data, and extracting a first number of fault features related to abnormal battery insulation failure based on the training set; constructing a random forest model based on the first number of fault features, and training the random forest model to obtain a trained random forest model.

[0008] In some embodiments of the present application, based on the aforementioned scheme, the method also includes a training step for a logistic regression model, and the training step of the logistic regression model includes: constructing a test set based on preprocessed historical battery national standard data; inputting a first number of fault features extracted based on the test set into the trained random forest model to output the importance probability of each fault feature through the trained random forest model; screening out a second number of fault features from the first number of fault features based on the importance probability of each fault feature; constructing a logistic regression model based on the second number of fault features, and adjusting the model parameters of the logistic regression model using a grid search method to obtain a trained logistic regression model.

[0009] In some embodiments of the present application, based on the aforementioned scheme, the penalty term coefficient of the trained logistic regression model is 11 or 12, the inverse of the regularization coefficient is 0.05, the ratio of normal vehicles to faulty vehicles is 0:0.99 or 1:0.01, and the random seed is 42.

[0010] In some embodiments of the present application, based on the aforementioned scheme, the first number of fault characteristics include the lowest insulation resistance continuously lower than 500KΩ, the number of frames in which the insulation resistance is continuously lower than 500KΩ, the duration in which the insulation resistance is continuously lower than 500KΩ, the average insulation resistance, the maximum insulation resistance, the operating status of the voltage converter, the number of continuous insulation alarm frames, the continuous alarm duration, the insulation alarm level, the average total voltage, the total voltage fluctuation pressure difference, the average current, the maximum voltage fluctuation pressure difference, the average voltage value, the charge and discharge status, the current change rate, the average temperature, and the temperature change rate; the second number of fault characteristics include the lowest insulation resistance continuously lower than 500KΩ, the number of frames in which the insulation resistance is continuously lower than 500KΩ, the duration in which the insulation resistance is continuously lower than 500KΩ, the average insulation resistance, the maximum insulation resistance, the operating status of the voltage converter, the number of continuous insulation alarm frames, the continuous alarm duration, the insulation alarm level, the average total voltage, the total voltage fluctuation pressure difference, the maximum voltage fluctuation pressure difference, the average voltage value, the charge and discharge status, and the current change rate.

[0011] In some embodiments of the present application, based on the aforementioned scheme, the warning results are screened again to ultimately determine the faulty vehicle with a battery insulation fault among the multiple vehicles to be detected, including: screening the warning results again to determine whether the insulation resistance of the vehicle to be detected whose warning result is an abnormality is lower than a preset insulation threshold for multiple consecutive frames; among the vehicles to be detected whose warning results are an abnormality, if there is any vehicle to be detected whose insulation resistance is lower than the preset insulation threshold for multiple consecutive frames, it is determined that the vehicle to be detected has a battery insulation fault.

[0012] In some embodiments of the present application, based on the aforementioned scheme, the fault types include charging pile insulation, whole vehicle insulation, and battery insulation, and determining the fault type of each faulty vehicle according to the charging status of each faulty vehicle and the operating status of the voltage converter includes: for any one faulty vehicle, when the faulty vehicle is in a charging state and the operating status of the voltage converter is not running, determining that the fault type of the faulty vehicle is charging pile insulation; for any one faulty vehicle, when the faulty vehicle is in a non-charging state and the operating status of the voltage converter is running, determining that the fault type of the faulty vehicle is whole vehicle insulation; for any one faulty vehicle, when the faulty vehicle is in an non-charging state and the operating status of the voltage converter is not running, determining that the fault type of the faulty vehicle is battery insulation.

[0013] According to the second aspect of an embodiment of the present application, a battery insulation abnormal fault detection device is provided, the device comprising: a data acquisition unit, used to obtain national standard battery data of multiple vehicles to be detected; a fault prediction unit, used to input the national standard battery data of multiple vehicles to be detected into a fault detection model to obtain a warning result of battery insulation failure for each vehicle to be detected, wherein the fault detection model comprises a pre-trained random forest model and a logistic regression model, and the warning result comprises the fault detection model; a verification unit, used to re-screen the warning result to ultimately determine a faulty vehicle with a battery insulation failure among the multiple vehicles to be detected; a type confirmation unit, used to determine the fault type of each faulty vehicle according to the charging status of each faulty vehicle and the operating status of the voltage converter.

[0014] According to a third aspect of an embodiment of the present application, a computer program product is provided, which includes computer instructions stored in a computer-readable storage medium and suitable for being read and executed by a processor, so that a computer device having the processor executes a method as described in any embodiment of the first aspect above.

[0015] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which at least one computer program instruction is stored. The at least one computer program instruction is loaded and executed by a processor to implement the operations performed by the method described in any embodiment of the first aspect above.

[0016] In this solution, the battery insulation abnormal fault warning algorithm driven by random forest and logistic regression data can achieve accurate warning of battery insulation abnormal faults through data collection and preprocessing, feature construction, random forest model construction, feature extraction, logistic regression model construction and fault warning. This solution has high warning accuracy, strong generalization ability and strong real-time performance for battery insulation abnormal faults. This solution is suitable for battery fault warning in the fields of electric vehicles and energy storage systems, and has high universality.

[0017] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0019] Figure 1 A flow chart showing a method for detecting abnormal insulation failure of a battery in an embodiment of the present application is shown;

[0020] Figure 2 A block diagram of a battery insulation abnormality fault detection device in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0022] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0023] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0024] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0025] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0026] Reference Figure 1 , shows a flow chart of a battery insulation abnormality fault detection method in an embodiment of the present application, and the battery insulation abnormality fault detection method can be executed by a device with computing and processing functions.

[0027] Reference Figure 1 As shown, the battery insulation abnormality fault detection method at least includes the following steps:

[0028] Step 110, obtaining national standard battery data of multiple vehicles to be tested.

[0029] Step 120, inputting the national standard battery data of multiple vehicles to be tested into the fault detection model to obtain a warning result of the battery insulation fault for each vehicle to be tested, wherein the fault detection model includes a pre-trained random forest model and a logistic regression model.

[0030] Step 130 , screening the warning results again to finally determine the faulty vehicle with battery insulation fault among the multiple vehicles to be detected.

[0031] Step 140 , determining the fault type of each faulty vehicle according to the charging state of each faulty vehicle and the operating state of the voltage converter.

[0032] First, the national standard battery data of multiple vehicles to be tested can be obtained through the cloud BMS. Among them, the national standard battery data includes battery voltage, current, temperature and other data. BMS refers to a battery management system, which is used to monitor and manage battery-related data. The number of vehicles to be tested can be multiple, which means that in this solution, the national standard battery data of multiple vehicles to be tested can be obtained at one time to uniformly perform fault detection operations for abnormal battery insulation. Furthermore, after obtaining the national standard battery data of multiple vehicles to be tested, the obtained data can be pre-processed by cleaning, denoising and other pre-processing operations. Then the pre-processed national standard battery data of multiple vehicles to be tested are input into the fault detection model. The fault detection model can output the early warning results of battery insulation faults for each vehicle to be tested based on the input data. Among them, the fault detection model integrates two models, namely a pre-trained random forest model and a logistic regression model. The random forest model refers to a classifier that uses multiple trees to train and predict samples. In machine learning, a random forest is a classifier containing multiple decision trees, and its output category is determined by the mode of the categories output by individual trees. The logistic regression model is a commonly used discriminant prediction model based on statistical methods.

[0033] After obtaining the early warning result of the battery insulation fault for each vehicle to be detected, it can be known whether each vehicle to be detected has a battery insulation fault according to the early warning result. Then, the early warning result can be screened again, so that the faulty vehicle with battery insulation fault among multiple vehicles to be detected can be finally determined to avoid false alarms. Furthermore, the fault type of each faulty vehicle can be determined according to the charging state of each faulty vehicle and the operating state of the voltage converter.

[0034] In one embodiment, the fault types of the vehicle include charging pile insulation, whole vehicle insulation, and battery insulation.

[0035] In one embodiment, the warning results are screened again to ultimately determine that a faulty vehicle has a battery insulation fault among multiple vehicles to be detected, including: screening the warning results again to determine whether the insulation resistance of the vehicle to be detected whose warning result is that there is an abnormality is lower than a preset insulation threshold for multiple consecutive frames; among the vehicles to be detected whose warning results are that there is an abnormality, if there is any vehicle to be detected whose insulation resistance is lower than the preset insulation threshold for multiple consecutive frames, it is determined that the vehicle to be detected has a battery insulation fault.

[0036] After determining the early warning result of the battery insulation fault for each vehicle to be detected based on the fault detection model, the early warning result can be screened again. Specifically, the screening method can be based on whether the insulation resistance of the vehicle to be detected is lower than the preset insulation threshold for multiple frames in a row. For any vehicle to be detected, if the early warning result of the vehicle to be detected is that there is an abnormality, it can be determined whether the insulation resistance of the vehicle to be detected is lower than the preset insulation threshold for multiple frames in a row. If its insulation resistance is lower than the preset insulation threshold for multiple frames in a row, it can be determined that the vehicle to be detected has a battery insulation fault problem. On the contrary, for any vehicle to be detected, if the early warning result of the vehicle to be detected is that there is an abnormality, and its insulation resistance is not lower than the preset insulation threshold for multiple frames in a row, it can be determined that the vehicle to be detected does not have a battery insulation fault problem. Specifically, the judgment condition can be set to 4 consecutive frames. That is, among the vehicles to be detected whose early warning results are abnormal, if there is any vehicle to be detected whose insulation resistance is lower than the preset insulation threshold for 4 consecutive frames, it can be determined that the vehicle to be detected has a battery insulation fault.

[0037] In one embodiment, the fault types of the vehicle include insulation of the charging pile, insulation of the whole vehicle, and insulation of the battery. Determining the fault type of each faulty vehicle according to the charging state of each faulty vehicle and the operating state of the voltage converter includes: for any faulty vehicle, when the faulty vehicle is in a charging state and the operating state of the voltage converter is not running, determining the fault type of the faulty vehicle is insulation of the charging pile; for any faulty vehicle, when the faulty vehicle is in a non-charging state and the operating state of the voltage converter is running, determining the fault type of the faulty vehicle is insulation of the whole vehicle; for any faulty vehicle, when the faulty vehicle is in a non-charging state and the operating state of the voltage converter is not running, determining the fault type of the faulty vehicle is insulation of the battery.

[0038] In the first case, for any faulty vehicle, if the faulty vehicle is in a charging state when the national standard battery data of the faulty vehicle is obtained, and the operating state of its voltage converter is not in operation, then it can be determined that the fault type of the faulty vehicle is insulation of the charging pile. Charging pile insulation usually means that a short circuit occurs in the high-voltage electrical components inside or related to the charging pile. The specific cause of the fault may include but is not limited to: damage or rust of internal parts of the charging pile; problems with the grounding system of the charging pile, such as poor grounding or poor grounding; failure of the insulation monitoring circuit or equipment of the charging pile, resulting in the inability to correctly detect insulation problems, etc. In the second case, for any faulty vehicle, if the faulty vehicle is in an uncharged state when the national standard battery data of the faulty vehicle is obtained, and the operating state of its voltage converter is in operation, then it can be determined that the fault type of the faulty vehicle is insulation of the whole vehicle. Among them, the insulation failure of the whole vehicle is caused by the leakage of electrolyte inside the battery, which causes liquid to seep out, thereby destroying the insulation layer, causing a conductive loop between the battery module and the monomer, thereby causing a fault. In the third case, for any faulty vehicle, if the faulty vehicle is in an uncharged state when the national standard battery data of the faulty vehicle is obtained, and the operating state of its voltage converter is not running, then it can be determined that the fault type of the faulty vehicle is battery insulation. Among them, the battery insulation failure is caused by the failure or damage of the insulating material inside the battery. Common reasons are as follows: 1. Battery aging causes aging of the insulating material and loss of insulation performance; 2. Damage to the battery shell causes the insulating shell to break or fall off, resulting in insulation failure; 3. The battery is left unused for a long time, resulting in internal corrosion of the battery and failure of the insulating material. Among them, the voltage converter can specifically refer to a DC-to-DC converter (DC-to-DC converter), also known as a DC-DC converter.

[0039] In one embodiment, the method also includes a training step for a random forest model, and the training step of the random forest model includes: collecting historical battery national standard data of multiple vehicles during historical operation; after preprocessing the historical battery national standard data, constructing a training set based on the preprocessed historical battery national standard data, and extracting a first number of fault features related to abnormal battery insulation failure based on the training set; constructing a random forest model based on the first number of fault features, and training the random forest model to obtain a trained random forest model.

[0040] In this embodiment, a training method for a random forest model is provided. First, historical battery national standard data of multiple vehicles in the historical operation process can be collected, that is, historical data of multiple vehicles are obtained. After preprocessing the historical battery national standard data, a training set can be constructed based on the preprocessed historical battery national standard data, and a first number of fault features related to abnormal battery insulation faults can be extracted based on the training set. Specifically, the first number can be defined by the technician. For example, in one embodiment, the first number of fault features may include the lowest insulation resistance continuously lower than 500KΩ, the number of frames in which the insulation resistance is continuously lower than 500KΩ, the duration in which the insulation resistance is continuously lower than 500KΩ, the average insulation resistance, the maximum insulation resistance, the operating state of the voltage converter, the number of continuous insulation alarm frames, the continuous alarm duration, the insulation alarm level, the average total voltage, the total voltage fluctuation pressure difference, the average current, the maximum voltage fluctuation pressure difference, the average voltage value, the charge and discharge state, the current change rate, the average temperature, the temperature change rate, etc. Then, a random forest model can be constructed based on the first number of fault features, and the random forest model can be trained to obtain a trained random forest model. The charge and discharge states include charge, discharge, pseudo-stationary and other states.

[0041] In one embodiment, the method further includes a training step for a logistic regression model. The training step of the logistic regression model includes: constructing a test set based on preprocessed historical battery national standard data; inputting a first number of fault features extracted based on the test set into a trained random forest model to output the importance probability of each fault feature through the trained random forest model; screening a second number of fault features from the first number of fault features based on the importance probability of each fault feature; constructing a logistic regression model based on the second number of fault features, and adjusting the model parameters of the logistic regression model using a grid search method to obtain a trained logistic regression model.

[0042] In this embodiment, a training method for a logistic regression model is provided. First, a test set can be constructed based on the preprocessed historical battery national standard data, and then the first number of fault features extracted based on the test set is input into the trained random forest model, and the trained random forest model can output the importance probability of each fault feature. Then the fault features can be sorted in order from high to low according to the numerical value of the importance probability, and the second number of fault features with higher importance probability can be screened out from the first number of fault features. For example, the first number is 18 and the second number is 15. Then, after the random forest model outputs the importance probability of each fault feature, 15 fault features with higher importance probability values ​​can be selected from the 18 fault features based on the order of importance probability from high to low. Then, a logistic regression model can be constructed based on the second number of fault features, and the model parameters of the logistic regression model can be adjusted using a grid search method to obtain a trained logistic regression model. When adjusting the model parameters of the logistic regression model using a grid search method, the scheme adopted by the prior art can be used, which will not be repeated here.

[0043] Among them, the second number of fault features may include the lowest insulation resistance continuously lower than 500KΩ, the number of frames in which the insulation resistance is continuously lower than 500KΩ, the duration in which the insulation resistance is continuously lower than 500KΩ, the average insulation resistance, the maximum insulation resistance, the operating state of the voltage converter, the number of continuous insulation alarm frames, the continuous alarm duration, the insulation alarm level, the average total voltage, the total voltage fluctuation pressure difference, the maximum voltage fluctuation pressure difference, the average voltage value, the charge and discharge state, and the current change rate. In a specific embodiment, the prediction accuracy of the logistic regression model trained based on this embodiment reaches 93.4%, and the recall rate is 92.8%.

[0044] In one embodiment, the penalty coefficient of the trained logistic regression model is 11 or 12, the inverse of the regularization coefficient is 0.05, the ratio of normal vehicles to faulty vehicles is 0:0.99 or 1:0.01, and the random seed is 42.

[0045] In this solution, the battery insulation abnormal fault warning algorithm driven by random forest and logistic regression data can achieve accurate warning of battery insulation abnormal faults through data collection and preprocessing, feature construction, random forest model construction, feature extraction, logistic regression model construction and fault warning. This solution has high warning accuracy, strong generalization ability and strong real-time performance for battery insulation abnormal faults. This solution is suitable for battery fault warning in the fields of electric vehicles and energy storage systems, and has high universality.

[0046] The following describes an apparatus embodiment of the present application, which can be used to execute the battery insulation abnormality fault detection method in the above-mentioned embodiment of the present application. For details not disclosed in the apparatus embodiment of the present application, please refer to the above-mentioned embodiment of the battery insulation abnormality fault detection method of the present application.

[0047] See also Figure 2 , shows a block diagram of a battery insulation abnormality fault detection device 200 in an embodiment of the present application. The device 200 includes: a data acquisition unit 201, a fault prediction unit 202, a verification unit 203 and a type confirmation unit 204.

[0048] Among them, the data acquisition unit 201 is used to obtain the national standard battery data of multiple vehicles to be detected; the fault prediction unit 202 is used to input the national standard battery data of multiple vehicles to be detected into the fault detection model to obtain the early warning result of the battery insulation fault for each vehicle to be detected, wherein the fault detection model includes a pre-trained random forest model and a logistic regression model, and the early warning result includes the fault detection model; the inspection unit 203 is used to re-screen the early warning result to finally determine the faulty vehicle with battery insulation fault among the multiple vehicles to be detected; and the type confirmation unit 204 is used to determine the fault type of each faulty vehicle according to the charging status of each faulty vehicle and the operating status of the voltage converter.

[0049] Based on the same inventive concept, an embodiment of the present application also provides a computer program product, which includes computer instructions, which are stored in a computer-readable storage medium and are suitable for being read and executed by a processor, so that a computer device having the processor executes the method as described above.

[0050] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, in which at least one computer program instruction is stored. The at least one computer program instruction is loaded and executed by a processor to implement the operations performed by the method described above.

[0051] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on a computer-readable medium or transmitted via a computer-readable medium. Other examples and implementations are within the scope and spirit of the present application and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hard wiring, or a combination of any of these. In addition, each functional unit may be integrated into a processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit.

[0052] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, 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 interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0053] The units described as separate components may or may not be physically separated, and the components of the control device may or may not be physical units, that is, they may be located in one place or distributed in multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0054] 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 storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc., which can store computer program instructions.

[0055] The above description is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. A battery insulation abnormality fault detection method, characterized in that: The method comprises: Obtain national standard battery data for multiple vehicles to be tested; Inputting national standard battery data of multiple vehicles to be tested into a fault detection model to obtain a warning result of battery insulation fault for each vehicle to be tested, wherein the fault detection model includes a pre-trained random forest model and a logistic regression model; The warning results are screened again to finally determine the faulty vehicle having the battery insulation fault among the multiple vehicles to be detected; The fault type of each faulty vehicle is determined based on the charging state of each faulty vehicle and the operating state of the voltage converter.

2. The method according to claim 1, characterized in that The method also includes a training step for a random forest model, and the training step of the random forest model includes: Collect historical battery national standard data of multiple vehicles during historical operation; After preprocessing the historical national standard data of the battery, a training set is constructed based on the preprocessed historical national standard data of the battery, and a first number of fault features related to abnormal battery insulation faults are extracted based on the training set; A random forest model is constructed based on the first number of fault features, and the random forest model is trained to obtain a trained random forest model.

3. The method according to claim 2, characterized in that The method further comprises a training step for a logistic regression model, wherein the training step for the logistic regression model comprises: Construct a test set based on preprocessed historical battery national standard data; Inputting a first number of fault features extracted based on the test set into the trained random forest model, so as to output the importance probability of each fault feature through the trained random forest model; Filtering a second number of fault features from the first number of fault features based on the importance probability of each fault feature; A logistic regression model is constructed based on the second number of fault features, and model parameters of the logistic regression model are adjusted using a grid search method to obtain a trained logistic regression model.

4. The method according to claim 3, characterized in that: The penalty coefficient of the trained logistic regression model is 11 or 12, the inverse of the regularization coefficient is 0.05, the ratio of normal vehicles to faulty vehicles is 0:0.99 or 1:0.01, and the random seed is 42.

5. The method according to any one of claims 2 to 4, characterized in that The first number of fault features includes the lowest insulation resistance continuously lower than 500KΩ, the number of frames in which the insulation resistance is continuously lower than 500KΩ, the duration in which the insulation resistance is continuously lower than 500KΩ, the average insulation resistance, the maximum insulation resistance, the operating state of the voltage converter, the number of continuous insulation alarm frames, the continuous alarm duration, the insulation alarm level, the average total voltage, the total voltage fluctuation pressure difference, the average current, the maximum voltage fluctuation pressure difference, the average voltage value, the charge and discharge state, the current change rate, the average temperature, and the temperature change rate; The second number of fault characteristics include the minimum insulation resistance continuously below 500KΩ, the number of frames when the insulation resistance is continuously below 500KΩ, the duration when the insulation resistance is continuously below 500KΩ, the average insulation resistance, the maximum insulation resistance, the operating status of the voltage converter, the number of continuous insulation alarm frames, the continuous alarm duration, the insulation alarm level, the average total voltage, the total voltage fluctuation pressure difference, the maximum voltage fluctuation pressure difference, the average voltage value, the charge and discharge status, and the current change rate.

6. The method according to claim 1, characterized in that The warning results are screened again to finally determine that the faulty vehicles with battery insulation faults among the multiple vehicles to be detected include: The warning result is screened again to determine whether the insulation resistance value of the vehicle to be detected, for which the warning result is abnormal, is lower than a preset insulation threshold value for multiple consecutive frames; If, among the vehicles to be detected for which the warning result is abnormal, the insulation resistance value of any one of the vehicles to be detected is lower than the preset insulation threshold for multiple consecutive frames, it is determined that the vehicle to be detected has a battery insulation fault.

7. The method according to claim 1, characterized in that The fault types include charging pile insulation, vehicle insulation, and battery insulation. The fault type of each faulty vehicle is determined according to the charging state of each faulty vehicle and the operating state of the voltage converter, including: For any faulty vehicle, when the faulty vehicle is in a charging state and the operating state of the voltage converter is not operating, determining that the fault type of the faulty vehicle is insulation of the charging pile; For any faulty vehicle, when the faulty vehicle is in an uncharged state and the operating state of the voltage converter is running, determining that the fault type of the faulty vehicle is whole vehicle insulation; For any faulty vehicle, when the faulty vehicle is in an uncharged state and the operating state of the voltage converter is not operating, it is determined that the fault type of the faulty vehicle is battery insulation.

8. A battery insulation abnormality fault detection device, characterized in that: The device comprises: A data acquisition unit, used to obtain national standard battery data of multiple vehicles to be tested; A fault prediction unit, used for inputting national standard battery data of multiple vehicles to be detected into a fault detection model to obtain a warning result of battery insulation fault for each vehicle to be detected, wherein the fault detection model includes a pre-trained random forest model and a logistic regression model, and the warning result includes the fault detection model; A testing unit, used for re-screening the warning result to finally determine the faulty vehicle having a battery insulation fault among the multiple vehicles to be tested; A type confirmation unit is used to determine the fault type of each faulty vehicle according to the charging state of each faulty vehicle and the operating state of the voltage converter.

9. A computer program product, characterized in that The computer program product includes computer instructions, which are stored in a computer-readable storage medium and are suitable for being read and executed by a processor, so that a computer device having the processor 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 computer program instructions, which are loaded and executed by a processor to implement the operations performed by the method according to any one of claims 1 to 7.

Images