Battery fault detection method and device, storage medium, electronic equipment and computer program product
By using a pre-trained target autoencoding model to process vehicle driving data and battery status data, the problems of complexity and inefficiency of battery failure detection in the prior art are solved, and efficient battery failure detection is achieved.
Patent Information
- Application Number
- CN202411797504.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-06
- Publication Date
- 2025-02-07
AI Technical Summary
The prior art requires the acquisition of a large amount of high-precision data in battery failure detection, which is complex in the processing process and low detection efficiency.
The pre-trained target self-coding model processes the travel data of the target vehicle and the status data of the target battery, outputs detection data, and determines the battery fault detection result based on the error information and the preset error threshold.
There is no need to build complex equivalent circuit models, which simplifies the data processing process, improves detection efficiency, and reduces the difficulty of training data acquisition.
Smart Images
Figure CN119395559A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of computer technology, and in particular to a battery fault detection method, device, storage medium, electronic device, and computer program product. Background Art
[0002] The battery is the most important component of the vehicle's power system and energy storage system, and its fault prediction and health management are crucial to the operation and maintenance of the vehicle. At present, the battery fault detection is usually carried out by building an equivalent circuit model. This method requires the collection of a large amount of high-precision battery data for model construction and calculation, and the processing process is complicated and the detection efficiency is low. Summary of the invention
[0003] In view of this, embodiments of the present application provide a battery fault detection method, device, storage medium, electronic device and computer program product to at least partially solve the above problems.
[0004] According to a first aspect of an embodiment of the present application, a battery fault detection method is provided, comprising: obtaining input data, the input data comprising driving data of a target vehicle and status data of a target battery in the target vehicle; processing the input data based on a pre-trained target autoencoder model, and outputting detection data, the target autoencoder model being a neural network model trained using normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle; determining error information between the detection data and the input data; and determining a fault detection result of the target battery based on the error information and a preset error threshold.
[0005] In some optional embodiments, the training process of the target autoencoder model includes: obtaining training data, the training data including normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle; based on the training data, performing unsupervised training on the autoencoder model to be trained to obtain the target autoencoder model.
[0006] In some optional embodiments, the method of this embodiment also includes: obtaining a test data set, the test data set including normal test data samples and abnormal test data samples, the normal test data samples including normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle, and the abnormal test data samples including abnormal driving data samples of the vehicle and abnormal status data samples of the battery in the vehicle; processing the test data set based on the target autoencoder model to obtain normal test data and abnormal test data; respectively determining first test error information of the normal test data samples and the normal test data, and second test error information of the abnormal test data samples and the abnormal test data; determining a preset error threshold based on the first test error information and the second test error information.
[0007] In some optional embodiments, the fault detection result of the target battery is determined based on the error information and the preset error threshold, including: if the error information is greater than or equal to the preset error threshold, the target battery is determined to be a faulty battery; if the error information is less than the preset error threshold, the target battery is determined to be a normal battery.
[0008] In some optional embodiments, the target autoencoder model includes an input layer, an output layer, and multiple hidden layers, and the number of the multiple hidden layers is less than or equal to 5.
[0009] In some optional embodiments, the multiple hidden layers include a first hidden layer, a second hidden layer, a third hidden layer, a fourth hidden layer and a fifth hidden layer; the number of neurons in the first hidden layer is twice the number of neurons in the second hidden layer, and the number of neurons in the second hidden layer is twice the number of neurons in the third hidden layer; the number of neurons in the first hidden layer is the same as the number of neurons in the fifth hidden layer, and the number of neurons in the second hidden layer is the same as the number of neurons in the fourth hidden layer.
[0010] According to a second aspect of an embodiment of the present application, a battery fault detection device is provided, comprising: an acquisition module for acquiring input data, the input data comprising driving data of a target vehicle and status data of a target battery in the target vehicle; a calculation module for processing the input data based on a pre-trained target autoencoder model and outputting detection data, the target autoencoder model being a neural network model trained by using normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle; an error determination module for determining error information between the detection data and the input data; and a detection result determination module for determining a fault detection result of the target battery based on the error information and a preset error threshold.
[0011] According to a third aspect of an embodiment of the present application, a computer storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method of the first aspect is implemented.
[0012] According to the fourth aspect of an embodiment of the present application, there is provided an electronic device, comprising a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; the memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform an operation corresponding to the method of the first aspect.
[0013] According to a fifth aspect of an embodiment of the present application, a computer program product is provided, comprising computer instructions, which instruct a computing device to perform operations corresponding to the method of the first aspect.
[0014] The embodiment of the present application provides a battery fault detection method, device, storage medium, electronic device and computer program product, wherein the battery fault detection method includes: obtaining input data, the input data includes the driving data of the target vehicle and the status data of the target battery in the target vehicle; processing the input data based on a pre-trained target autoencoder model and outputting detection data, the target autoencoder model is a neural network model trained by normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle; determining the error information between the detection data and the input data; and determining the fault detection result of the target battery according to the error information and a preset error threshold. The embodiment of the present application processes the driving data of the target vehicle and the status data of the target battery in the target vehicle through a pre-trained target autoencoder model to obtain detection data, and determines the fault detection result of the target battery based on the error information between the detection data and the input data and a preset error threshold, without the need to construct a complex equivalent circuit model, thereby improving the detection efficiency; and the target autoencoder model can be trained through normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle, and the difficulty of collecting training data is relatively low. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0016] Figure 1 is a flowchart of a battery failure detection method according to an exemplary embodiment of the present application;
[0017] Figure 2 is a schematic diagram of a target autoencoder model of a battery fault detection method according to an exemplary embodiment of the present application;
[0018] Figure 3 is a structural block diagram of a battery fault detection device according to an exemplary embodiment of the present application;
[0019] Figure 4 It is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present application. DETAILED DESCRIPTION
[0020] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings 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. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the embodiments of the present application should fall within the scope of protection of the embodiments of the present application.
[0021] Reference Figure 1 , shows a step flow chart of a battery fault detection method according to an exemplary embodiment of the present application.
[0022] As shown in the figure, this embodiment mainly includes the following steps:
[0023] S102: Obtain input data.
[0024] The battery fault detection method of this embodiment can be applied to user equipment, and the user equipment may include any one or more user equipment suitable for displaying the fault detection result of the battery, interacting with the user, etc. In some embodiments, the user equipment may include any suitable type of equipment. For example, in some embodiments, the user equipment may include a mobile device, a tablet computer, a laptop computer, a desktop computer, and / or any other suitable type of user equipment.
[0025] The input data includes the driving data of the target vehicle and the state data of the target battery in the target vehicle. Exemplarily, the driving data of the target vehicle and the state data of the target battery in the target vehicle can both be time series data, and the driving data of the target vehicle refers to data that can reflect the battery state, for example, it can include vehicle power state data, etc. The state data of the target battery can include battery current, voltage, SOC, battery temperature, SOH, battery internal resistance, battery capacity, battery charging mode, etc., wherein SOC refers to the state of charge of the battery, that is, the ratio between the current charge level stored in the battery and its maximum charging capacity. SOC can reflect how much available electrical energy is left in the battery. SOH refers to the health state of the battery, that is, the degree of performance degradation of the battery during use. SOH can reflect the percentage of the remaining capacity of the battery relative to its initial design capacity. The driving data of the target vehicle and the state data of the target battery in the target vehicle can form a multidimensional matrix, and the input data is input into the target autoencoder model in the form of a multidimensional matrix.
[0026] S104: Process the input data based on the pre-trained target autoencoder model and output detection data.
[0027] Among them, the target autoencoder model is a neural network model trained by normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle.
[0028] Exemplarily, the target autoencoder model can be pre-trained, and the training data includes normal driving data samples of the vehicle and normal state data samples of the battery in the vehicle. The normal driving data samples can be historical normal driving data of multiple vehicles, and the normal state data samples can be historical normal state data of the battery in multiple vehicles. The autoencoder model is trained using the training data to obtain the target autoencoder model. The target autoencoder model can first encode the input data, and then decode the encoded input data, that is, reconstruct and restore the encoded input data, and output it. Since the target autoencoder model is trained using normal driving data samples of the vehicle and normal state data samples of the battery in the vehicle, when the input data is normal driving data and normal state data, the output detection data is basically consistent with the input data; when the input data is abnormal driving data and / or normal state data, it is difficult for the target autoencoder model to restore the input data, and the error between the output detection data and the input data is large.
[0029] In some optional embodiments, the training process of the target autoencoder model includes: obtaining training data, the training data including normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle; based on the training data, performing unsupervised training on the autoencoder model to be trained to obtain the target autoencoder model.
[0030] Exemplarily, the normal driving data samples may be historical normal driving data of multiple vehicles, and the normal status data samples may be historical normal status data of batteries in multiple vehicles. For example, historical normal driving data of 1 month and historical normal status data of batteries in vehicles may be intercepted for 10,000 vehicles respectively, and the data may be sliced by hour, so that 7.2 million data segments may be obtained, which may be used as training data sets. The autoencoder model may be unsupervisedly trained using the training data. For example, an autoencoder model may be constructed. Assuming that the amount of data in the training data set is 6.48 million training data, the number of cycles (i.e., epochs) of the entire training data set is set to 50 times, and the amount of data input for a single training (batchsize) is set to 32 training data. The autoencoder model may be trained to obtain the target autoencoder model.
[0031] In this embodiment, the training process of the target autoencoder model is an unsupervised training process, and there is no need to collect labeled data, thereby reducing the amount and difficulty of data collection for model training.
[0032] In some optional embodiments, the target autoencoder model includes an input layer, an output layer, and multiple hidden layers, and the number of the multiple hidden layers is less than or equal to 5.
[0033] In this embodiment, the target autoencoder model may include an input layer, an output layer, and multiple hidden layers. The multiple hidden layers may perform multi-level feature extraction on the input data, and ultimately better linearly divide different types of data features in the input data, thereby more accurately determining the rules between different types of data. The number of the multiple hidden layers is less than or equal to 5. Preferably, the target autoencoder model may include 5 hidden layers, that is, the target autoencoder model of this embodiment has a total of 7 layers. For example, referring to Figure 2 , the target autoencoder model includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer, a fourth hidden layer, a fifth hidden layer and an output layer.
[0034] In this embodiment, by setting multiple hidden layers, the input data can be used for more accurate feature extraction and reconstruction and restoration, thereby improving the accuracy of the detection results.
[0035] In some optional embodiments, the multiple hidden layers include a first hidden layer, a second hidden layer, a third hidden layer, a fourth hidden layer and a fifth hidden layer; the number of neurons in the first hidden layer is twice the number of neurons in the second hidden layer, and the number of neurons in the second hidden layer is twice the number of neurons in the third hidden layer; the number of neurons in the first hidden layer is the same as the number of neurons in the fifth hidden layer, and the number of neurons in the second hidden layer is the same as the number of neurons in the fourth hidden layer.
[0036] For example, refer to Figure 2 , the target autoencoder model of this embodiment may include 5 hidden layers, wherein the number of neurons in the first hidden layer is twice the number of neurons in the second hidden layer, for example, the number of neurons in the first hidden layer may be 16, and the number of neurons in the second hidden layer may be 8. The number of neurons in the second hidden layer is twice the number of neurons in the third hidden layer, for example, the number of neurons in the third hidden layer may be 4. Moreover, the number of neurons in the first hidden layer is the same as the number of neurons in the fifth hidden layer, and the number of neurons in the second hidden layer is the same as the number of neurons in the fourth hidden layer, for example, the number of neurons in the fourth hidden layer may be 8, and the number of neurons in the fifth hidden layer may be 16.
[0037] In this embodiment, the target autoencoder model is set to include 5 hidden layers, and the number of neurons in the first hidden layer is twice the number of neurons in the second hidden layer, and the number of neurons in the second hidden layer is twice the number of neurons in the third hidden layer; the number of neurons in the first hidden layer is the same as the number of neurons in the fifth hidden layer, and the number of neurons in the second hidden layer is the same as the number of neurons in the fourth hidden layer, avoiding the problem of too many parameters to be trained due to too many hidden layers, which in turn leads to reduced model training efficiency.
[0038] S106: Determine error information between the detection data and the input data.
[0039] Exemplarily, the detection data is compared with the input data, and the error value between the detection data and the input data is calculated. Taking the case where both the detection data and the input data are multidimensional matrices, each element value in the detection data is subtracted from the corresponding element value in the input data to obtain a difference matrix, and then the norm of the difference matrix is calculated to obtain error information, which can be a numerical value.
[0040] S108. Determine a fault detection result of the target battery according to the error information and a preset error threshold.
[0041] Exemplarily, the determined error information is compared with a preset error threshold to determine whether the error information is greater than the preset error threshold, thereby determining a fault detection result of the target battery according to the determination result.
[0042] Among them, the preset error threshold can be obtained in the following manner: obtain a test data set, the test data set includes normal test data samples and abnormal test data samples, the normal test data samples include normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle, and the abnormal test data samples include abnormal driving data samples of the vehicle and abnormal status data samples of the battery in the vehicle; based on the target autoencoder model, process the test data set to obtain normal test data and abnormal test data; respectively determine the first test error information of the normal test data sample and the normal test data, and the second test error information of the abnormal test data sample and the abnormal test data; determine the preset error threshold according to the first test error information and the second test error information.
[0043] Exemplarily, historical driving data of multiple vehicles and historical status data of batteries in the vehicles that are different from the training data set can be obtained as a test data set, and the test data set can include normal test data samples and abnormal test data samples, wherein the normal test data samples include normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle, and the abnormal test data samples include abnormal driving data samples of the vehicle and abnormal status data samples of the battery in the vehicle. The data in the test data set is input into the target autoencoder model for inference, and normal test data corresponding to the normal test data samples and abnormal test data corresponding to the abnormal test data samples can be obtained. Afterwards, the first test error information between the normal test data sample and the normal test data can be determined, and the second test error information between the abnormal test data sample and the abnormal test data can be determined. The obtained first test error information and the second test error information are statistically analyzed to determine a preset error threshold, and the normal test data sample and the abnormal test data sample can be distinguished by the preset error threshold.
[0044] In this embodiment, a test data set including normal test data samples and abnormal test data samples is set, and the corresponding normal test data and abnormal test data are obtained by inferring the data in the test data set using the trained target autoencoder model, and then the first test error information of the normal test data samples and the normal test data, as well as the second test error information of the abnormal test data samples and the abnormal test data are respectively determined, and then the preset error threshold is determined based on the first test error information and the second test error information, so as to improve the accuracy of the error information judgment between the input data and the detection data.
[0045] In some optional embodiments, the fault detection result of the target battery is determined based on the error information and the preset error threshold, including: if the error information is greater than or equal to the preset error threshold, the target battery is determined to be a faulty battery; if the error information is less than the preset error threshold, the target battery is determined to be a normal battery.
[0046] Exemplarily, the determined error information is compared with a preset error threshold, and if the error information is greater than or equal to the preset error threshold, it indicates that the fault detection result is that the target battery is a faulty battery. If the error information is less than the preset error threshold, it indicates that the fault detection result is that the target battery is a normal battery.
[0047] In this embodiment, the fault detection result of the target battery can be determined according to the magnitude relationship between the error information and the preset error threshold, and the detection process is simple and the detection efficiency is high.
[0048] The embodiment of the present application processes the driving data of the target vehicle and the status data of the target battery in the target vehicle through a pre-trained target autoencoder model to obtain detection data, and determines the fault detection result of the target battery based on the error information between the detection data and the input data and a preset error threshold. There is no need to construct a complex equivalent circuit model, thereby improving the detection efficiency; and the target autoencoder model can be trained through normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle, and the difficulty of collecting training data is relatively low.
[0049] Reference Figure 3 , shows a structural block diagram of a battery fault detection device according to an exemplary embodiment of the present application.
[0050] The vehicle simulation test device of this embodiment includes an acquisition module 302 , a calculation module 304 , an error determination module 306 and a detection result determination module 308 .
[0051] Among them, the acquisition module 302 is used to acquire input data, and the input data includes the driving data of the target vehicle and the status data of the target battery in the target vehicle; the calculation module 304 is used to process the input data based on the pre-trained target autoencoder model and output the detection data, and the target autoencoder model is a neural network model trained by the normal driving data samples of the vehicle and the normal status data samples of the battery in the vehicle; the error determination module 306 is used to determine the error information between the detection data and the input data; the detection result determination module 308 is used to determine the fault detection result of the target battery according to the error information and the preset error threshold.
[0052] In some optional embodiments, the device of this embodiment also includes a training module, which is used to: obtain training data, the training data including normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle; based on the training data, perform unsupervised training on the autoencoder model to be trained to obtain a target autoencoder model.
[0053] In some optional embodiments, the device of this embodiment also includes a preprocessing module, which is used to: obtain a test data set, the test data set including normal test data samples and abnormal test data samples, the normal test data samples including normal driving data samples of the vehicle and normal status data samples of the battery in the vehicle, and the abnormal test data samples including abnormal driving data samples of the vehicle and abnormal status data samples of the battery in the vehicle; based on the target autoencoder model, process the test data set to obtain normal test data and abnormal test data; respectively determine first test error information of the normal test data sample and the normal test data, and second test error information of the abnormal test data sample and the abnormal test data; determine a preset error threshold based on the first test error information and the second test error information.
[0054] In some optional embodiments, the detection result determination module 308 is further used to: determine that the target battery is a faulty battery if the error information is greater than or equal to a preset error threshold; and determine that the target battery is a normal battery if the error information is less than the preset error threshold.
[0055] In some optional embodiments, the target autoencoder model includes an input layer, an output layer, and multiple hidden layers, and the number of the multiple hidden layers is less than or equal to 5.
[0056] In some optional embodiments, the multiple hidden layers include a first hidden layer, a second hidden layer, a third hidden layer, a fourth hidden layer and a fifth hidden layer; the number of neurons in the first hidden layer is twice the number of neurons in the second hidden layer, and the number of neurons in the second hidden layer is twice the number of neurons in the third hidden layer; the number of neurons in the first hidden layer is the same as the number of neurons in the fifth hidden layer, and the number of neurons in the second hidden layer is the same as the number of neurons in the fourth hidden layer.
[0057] The battery fault detection device of this embodiment is used to implement the corresponding battery fault detection methods in the aforementioned multiple method embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be described in detail here. In addition, the functional implementation of each module in the battery fault detection device of this embodiment can refer to the description of the corresponding parts in the aforementioned method embodiments, which will not be described in detail here.
[0058] Reference Figure 4 , shows a schematic diagram of the structure of an electronic device according to an exemplary embodiment of the present application. The specific embodiment of the present invention does not limit the specific implementation of the electronic device.
[0059] like Figure 4 As shown, the electronic device may include: a processor (processor) 401 , a memory (memory) 403 , a communication bus 404 , and a communication interface (Communications Interface) 405 .
[0060] Wherein: the processor 401 , the memory 403 and the communication interface 405 communicate with each other via the communication bus 404 .
[0061] The communication interface 405 is used to communicate with other electronic devices or servers.
[0062] The processor 401 is used to execute the program 402, and specifically can execute the steps of any one of the battery failure detection methods in the above embodiments.
[0063] Specifically, the program 402 may include program codes, which include computer operation instructions.
[0064] The processor 401 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the smart device may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.
[0065] The memory 403 is used to store the program 402. The memory 403 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
[0066] The program 402 can be specifically used to enable the processor 401 to execute to implement the steps of any battery fault detection method described in the embodiment. The specific implementation of each step in the program 402 can refer to the corresponding description of the steps and units executed by any battery fault detection method in the above steps, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described devices and modules can refer to the corresponding process description in the above-mentioned method embodiment.
[0067] An embodiment of the present application further provides a computer storage medium on which a computer program is stored. When the program is executed by a processor, the battery fault detection method as described in any one of the above-mentioned method embodiments is implemented.
[0068] An embodiment of the present application further provides a computer program product, including computer instructions, which instruct a computing device to execute operations corresponding to the battery fault detection method described in any one of the above-mentioned method embodiments.
[0069] It should be pointed out that, according to the needs of implementation, the various components / steps described in the embodiments of the present application can be split into more components / steps, or two or more components / steps or partial operations of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of the present application.
[0070] The above-mentioned method according to the embodiment of the present application can be implemented in hardware, firmware, or implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk or magneto-optical disk), or implemented as a computer code originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded through a network and to be stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a special-purpose processor or programmable or special-purpose hardware (such as an ASIC or FPGA). It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, a processor or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown here, the execution of the code converts the general-purpose computer into a special-purpose computer for executing the method shown here.
[0071] Those of ordinary skill in the art will appreciate that the units and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the embodiments of the present application.
[0072] The above implementation methods are only used to illustrate the embodiments of the present application, and are not limitations on the embodiments of the present application. Ordinary technicians in the relevant technical field can make various changes and modifications without departing from the spirit and scope of the embodiments of the present application. Therefore, all equivalent technical solutions also belong to the scope of the embodiments of the present application. The scope of patent protection of the embodiments of the present application should be limited by the claims.
Claims
1. A battery failure detection method, characterized in that: include: Acquiring input data, the input data including driving data of a target vehicle and status data of a target battery in the target vehicle; Processing the input data based on a pre-trained target autoencoder model and outputting detection data, wherein the target autoencoder model is a neural network model trained by a normal driving data sample of the vehicle and a normal state data sample of the battery in the vehicle; Determining error information between the detection data and the input data; A fault detection result of the target battery is determined according to the error information and a preset error threshold.
2. The battery failure detection method according to claim 1, characterized in that: The training process of the target autoencoder model includes: Acquire training data, the training data comprising a normal driving data sample of a vehicle and a normal state data sample of a battery in the vehicle; Based on the training data, unsupervised training is performed on the autoencoder model to be trained to obtain a target autoencoder model.
3. The battery failure detection method according to claim 2, characterized in that: The method further comprises: Acquire a test data set, the test data set comprising normal test data samples and abnormal test data samples, the normal test data samples comprising normal driving data samples of the vehicle and normal state data samples of the battery in the vehicle, and the abnormal test data samples comprising abnormal driving data samples of the vehicle and abnormal state data samples of the battery in the vehicle; Based on the target autoencoder model, the test data set is processed to obtain normal test data and abnormal test data; respectively determining the normal test data sample and first test error information of the normal test data, and the abnormal test data sample and second test error information of the abnormal test data; The preset error threshold is determined according to the first test error information and the second test error information.
4. The battery failure detection method according to claim 1, characterized in that: The step of determining the fault detection result of the target battery according to the error information and a preset error threshold comprises: If the error information is greater than or equal to a preset error threshold, determining that the target battery is a faulty battery; If the error information is less than the preset error threshold, it is determined that the target battery is a normal battery.
5. The battery failure detection method according to claim 1, characterized in that: The target autoencoder model includes an input layer, an output layer and multiple hidden layers, and the number of the multiple hidden layers is less than or equal to 5.
6. The battery failure detection method according to claim 5, characterized in that: The plurality of hidden layers include a first hidden layer, a second hidden layer, a third hidden layer, a fourth hidden layer and a fifth hidden layer; The number of neurons in the first hidden layer is twice the number of neurons in the second hidden layer, and the number of neurons in the second hidden layer is twice the number of neurons in the third hidden layer; The number of neurons in the first hidden layer is the same as the number of neurons in the fifth hidden layer, and the number of neurons in the second hidden layer is the same as the number of neurons in the fourth hidden layer.
7. A battery fault detection device, characterized in that: include: An acquisition module, used for acquiring input data, wherein the input data includes driving data of a target vehicle and status data of a target battery in the target vehicle; A computing module, used to process the input data based on a pre-trained target autoencoder model and output detection data, wherein the target autoencoder model is a neural network model trained by a normal driving data sample of the vehicle and a normal state data sample of the battery in the vehicle; An error determination module, used to determine error information between the detection data and the input data; The detection result determination module is used to determine the fault detection result of the target battery according to the error information and a preset error threshold.
8. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform an operation corresponding to the method according to any one of claims 1 to 6.
9. A computer storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.
10. A computer program product, characterized in that The method comprises computer instructions, wherein the computer instructions instruct a computing device to execute operations corresponding to the method according to any one of claims 1 to 6.