Battery fault detection method and device and nonvolatile storage medium

Abstract

The invention discloses a battery fault detection method and device and a nonvolatile storage medium. The method comprises the following steps: collecting battery data of a battery of a target vehicle during charging through a charging pile for charging the target vehicle; determining the type of the battery of the target vehicle according to the battery data; the type of the battery and the battery data are input into a pre-trained neural network model, the neural network model outputs a detection result of whether the battery has a fault or not, the neural network model is obtained by training an original neural network model through a training sample, the training sample comprises sample battery data, a sample battery type and a fault tag, and the fault tag is a fault tag; the sample battery data is obtained by performing data acquisition on the sample vehicle through the sample charging pile. The technical problem that a battery fault detection method in the prior art cannot be applied to a charging end is solved.

Description

Technical Field

[0001] The present invention relates to the field of new energy technologies, and in particular to a battery fault detection method, device and non-volatile storage medium. Background Art

[0002] According to statistics, more than 50% of new energy vehicle fires occur within 10 to 24 hours after charging, indicating that there is a correlation between new energy vehicle charging behavior and fire accidents. The existing technology for battery safety assessment mainly focuses on the cloud big data methods of vehicle companies, relying on the extraction of single vehicle features and judging whether it has a fault. There are few research technologies for battery safety warning under the condition that the charging pile lacks information such as the voltage and temperature of all single cells.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a battery fault detection method, device and non-volatile storage medium to at least solve the technical problem that the battery fault detection method in the related art cannot be applied to the charging end.

[0005] According to one aspect of an embodiment of the present invention, a battery fault detection method is provided, comprising: collecting battery data of a battery of a target vehicle during charging through a charging pile for charging the target vehicle; determining the type of the battery of the target vehicle based on the battery data; inputting the battery type and battery data into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training an original neural network model through training samples, the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data on a sample vehicle through a sample charging pile.

[0006] Optionally, the battery type and battery data are input into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, including: determining historical charging characteristics through historical data recorded by the charging pile, wherein the historical charging characteristics include at least one of the following: historical fast charging times, historical charging depth, historical charging time, and cumulative charge and discharge capacity; combining the historical charging characteristics, battery type, and battery data, and inputting the combined data into the neural network model, and the neural network model outputs the detection result.

[0007] Optionally, when the battery data includes a battery state of charge, determining the type of the battery of the target vehicle based on the battery data includes: respectively obtaining multiple battery states of charge of the target vehicle at multiple moments; combining the multiple battery states of charge with the multiple moments in a one-to-one correspondence to obtain multiple two-dimensional data points; and determining the type of the battery based on the multiple two-dimensional data points.

[0008] Optionally, determining the type of battery based on multiple two-dimensional data points includes: obtaining multiple sample two-dimensional data points, wherein the types of batteries corresponding to the multiple sample two-dimensional data points have been determined, and the types of batteries corresponding to the multiple sample two-dimensional data points include lithium iron phosphate batteries and other batteries; clustering the multiple sample two-dimensional data points and the multiple two-dimensional data points to obtain multiple classifications; and determining the type of battery based on the multiple classifications.

[0009] Optionally, the training samples are determined in the following manner: based on the sample charging pile, sample battery data of the sample vehicle is obtained; the sample battery data is classified to determine the sample battery type; the fault condition of the sample vehicle is obtained to determine the fault label corresponding to the sample battery data; and training samples are generated based on the sample battery data, sample battery type and fault label.

[0010] Optionally, the neural network model is a random decision forest model.

[0011] According to another aspect of an embodiment of the present invention, a battery fault detection device is also provided, including: an acquisition module, used to acquire battery data of a battery of a target vehicle during charging through a charging pile for charging the target vehicle; a determination module, used to determine the type of battery of the target vehicle based on the battery data; a detection module, used to input the battery type and battery data into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training an original neural network model through training samples, the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data from a sample vehicle through a sample charging pile.

[0012] According to another aspect of an embodiment of the present invention, a non-volatile storage medium is provided, the non-volatile storage medium including a stored program, wherein when the program is running, the device where the non-volatile storage medium is located is controlled to execute any one of the above-mentioned battery fault detection methods.

[0013] According to another aspect of the embodiments of the present invention, a computer device is provided. The computer device includes a processor, and the processor is used to run a program. When the program is run, any one of the above-mentioned battery failure detection methods is executed.

[0014] According to another aspect of the embodiments of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, any one of the above-mentioned battery fault detection methods is implemented.

[0015] In an embodiment of the present invention, battery data of the battery of the target vehicle during charging is collected through a charging pile for charging a target vehicle; the type of the battery of the target vehicle is determined based on the battery data; the battery type and battery data are input into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training an original neural network model through training samples, and the training samples include sample battery data, sample battery types and fault labels, and the sample battery data is obtained by collecting data from sample vehicles through sample charging piles, thereby achieving the purpose of performing battery fault detection at the charging end based on relevant data of the charging vehicle, thereby achieving the technical effect of detecting battery faults when charging electric vehicles at the charging end, and further solving the technical problem that the battery fault detection method in the related art cannot be applied to the charging end. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0017] Figure 1 A hardware structure block diagram of a computer terminal for implementing a battery failure detection method is shown;

[0018] Figure 2 is a flow chart of a battery fault detection method provided according to an embodiment of the present invention;

[0019] Figure 3 is a schematic diagram of a neural network model training process provided according to an optional embodiment of the present invention;

[0020] Figure 4 is a structural block diagram of a battery fault detection device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] According to an embodiment of the present invention, a method embodiment of battery fault detection is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0024] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal for implementing a battery failure detection method. Figure 1 As shown, the computer terminal 10 may include one or more (102a, 102b, ..., 102n are used to illustrate) processors (the processor may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0025] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10. As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0026] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the battery fault detection method in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the battery fault detection method of the above-mentioned application is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0027] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0028] Since the charging action of electric vehicles is closely related to the cause of battery failure, it is of great application significance to evaluate the safety of new energy vehicles from the charging side. The main cause of fire in new energy vehicles is the power battery, which can easily cause thermal runaway of the battery due to internal short circuit, lithium deposition, overcharging, etc., and then spread to the whole vehicle fire. Therefore, analyzing the safety of the battery from the charging end data, screening out abnormal batteries and issuing alarms can detect safety problems of new energy vehicles earlier and deal with them in an emergency, which can reduce the harm of new energy vehicle fire accidents and help protect the assets of the charging pile end.

[0029] Figure 2 FIG. 1 is a flow chart of a battery fault detection method according to an embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0030] Step S202, collecting battery data of the battery of the target vehicle during charging through the charging pile for charging the target vehicle.

[0031] In this step, the charging pile will monitor the data of the battery charging process in real time by connecting to the battery system of the target vehicle, such as voltage, current, temperature and other parameters. The charging pile can also record its own charging voltage and current data in real time, which will be used for subsequent analysis and processing.

[0032] Step S204, determining the type of the battery of the target vehicle according to the battery data.

[0033] In this step, by analyzing the collected battery data, the type of battery used by the target vehicle can be determined, such as lithium battery, nickel-metal hydride battery, etc. Because different types of batteries may have different data performance, the battery type is very important for subsequent fault detection and can effectively help fault detection.

[0034] Step S206, input the battery type and battery data into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training the original neural network model through training samples, and the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data from a sample vehicle through a sample charging pile.

[0035] In this step, the determined battery type and the collected battery data are input into a pre-trained neural network model for processing. The neural network model has learned the characteristics of battery failure through previous training samples, and can make a judgment on whether the battery is faulty based on the input data. In this way, the battery fault can be detected in real time at the charging end, and potential problems can be discovered and handled in advance.

[0036] Through the above steps, the purpose of performing battery fault detection at the charging end according to the relevant data of the charging vehicle can be achieved, thereby realizing the technical effect of detecting battery faults when charging the electric vehicle at the charging end, thereby solving the technical problem that the battery fault detection method in the related technology cannot be applied to the charging end.

[0037] As an optional embodiment, the battery type and battery data are input into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, including: determining historical charging characteristics through historical data recorded by the charging pile, wherein the historical charging characteristics include at least one of the following: historical fast charging times, historical charging depth, historical charging time, and cumulative charge and discharge capacity; combining the historical charging characteristics, battery type, and battery data, and inputting the combined data into the neural network model, and the neural network model outputs the detection result.

[0038] Optionally, this optional embodiment may combine the battery type and battery data with historical charging characteristics, and use a pre-trained neural network model to detect battery faults. Historical charging characteristics may include information such as historical fast charge times, historical charging depth, historical charging time, and cumulative charge and discharge capacity, which may reflect the usage and health of the battery. By combining these features with the battery data and inputting them into the neural network model, it is possible to more accurately determine whether the battery is faulty. Historical features can be features that can effectively determine battery faults after professionals analyze the data. The neural network model can comprehensively predict the battery status based on these features and the features it has learned, thereby improving the accuracy of battery fault detection.

[0039] Among them, the number of fast charges reflects the number of times the battery is charged under high power conditions. The number of fast charges of the battery can be counted through the charging pile; the charging depth reflects the change in SOC of the battery at the end of each charge, which can be compared by comparing the change in SOC of the battery before and after charging; when the battery has problems such as capacity diving, the battery SOC change will increase under the same charging ampere-hour conditions, or the cumulative ampere-hours will become smaller under the same SOC change. Therefore, the cumulative charge and discharge amount can also reflect whether the battery has a fault. Specifically, it can be based on all historical data of a certain vehicle, with the lowest SOC span of all charging segments or the specified SOC span as the outline, and the cumulative ampere-hours of the battery in a specific SOC range can be calculated, or the SOC change under the same charging ampere-hour conditions can be calculated to obtain the statistical distribution of the calculated change, and judge whether the battery is faulty based on the distribution. Specifically, the 3σ principle can be used for hypothesis testing, and an alarm can be issued for charging segments exceeding 3σ.

[0040] As an optional embodiment, when the battery data includes the battery state of charge, determining the type of the battery of the target vehicle based on the battery data includes: respectively obtaining multiple battery states of charge of the target vehicle at multiple moments; combining the multiple battery states of charge with the multiple moments in a one-to-one correspondence to obtain multiple two-dimensional data points; and determining the type of the battery based on the multiple two-dimensional data points.

[0041] Optionally, the type of battery used by the target vehicle can be determined by analyzing the battery state of charge data of the target vehicle at different times. First, it is necessary to obtain the battery state of charge data of the target vehicle at multiple times, that is, the SOC value of the battery. Then, these battery state of charge data are combined with the corresponding time to obtain multiple two-dimensional data points, each of which contains the battery state of charge information at a time. By analyzing and processing these two-dimensional data points, the type of battery can be determined.

[0042] Among them, the SOC value (State of Charge) of the battery represents the percentage of electrical energy stored in the battery relative to its maximum capacity. SOC is a dimensionless value, usually in the range of 0 to 100%. When the SOC is 0%, the battery is fully discharged; when the SOC is 100%, the battery is fully charged. The SOC value is very important for understanding the remaining power of the battery and estimating the endurance.

[0043] As an optional embodiment, determining the type of a battery based on multiple two-dimensional data points includes: obtaining multiple sample two-dimensional data points, wherein the types of batteries corresponding to the multiple sample two-dimensional data points have been determined, and the types of batteries corresponding to the multiple sample two-dimensional data points include lithium iron phosphate batteries and other batteries; clustering the multiple sample two-dimensional data points and the multiple two-dimensional data points to obtain multiple classifications; and determining the type of the battery based on the multiple classifications.

[0044] Optionally, a clustering algorithm or classification algorithm can be used to classify data points to determine that batteries with similar state of charge data belong to the same type. In this way, the type of battery used in the target vehicle can be accurately determined, providing an important reference for subsequent battery management and maintenance. This method can help vehicle manufacturers or owners better understand the use of batteries and take corresponding maintenance and management measures in a timely manner.

[0045] Optionally, the two-dimensional data points can also be determined by the battery's state of charge data and the battery's open circuit voltage data, that is, the battery's OCV value (Open Circuit Voltage), which is the voltage of the battery when no load is connected. The OCV value changes with the battery's SOC, and is usually higher when the battery is close to being fully charged, and lower when it is close to being fully discharged. By measuring the OCV value, the battery's SOC value and health status can be estimated. Specifically, the battery's internal resistance can be extracted based on the step voltage at the end of charging or during charging, and then the battery's internal resistance is assumed to be unchanged. The OCV at this point is calculated by subtracting the terminal voltage from the current multiplied by the internal resistance value, and the OCV-SOC curve under the charging segment is obtained, thereby obtaining multiple two-dimensional data points. At this point, based on the characteristics of lithium iron phosphate batteries, in the plane where the SOC-OCV sample points are located, the cluster with a lower OCV can be calibrated as a lithium iron phosphate battery.

[0046] As an optional embodiment, the training samples are determined in the following manner: based on the sample charging pile, the sample battery data of the sample vehicle is obtained; the sample battery data is classified to determine the sample battery type; the fault condition of the sample vehicle is obtained to determine the fault label corresponding to the sample battery data; and the training samples are generated according to the sample battery data, the sample battery type and the fault label.

[0047] Optionally, the battery data of the sample vehicle can be obtained based on the sample charging pile. These data may include information such as the battery's state of charge, number of charges, and depth of charge. Then, these sample battery data are classified to determine the type of battery used by the sample vehicle. Next, the fault condition of the sample vehicle is obtained, and the fault label corresponding to the sample battery data is determined, that is, which sample battery data corresponds to the faulty battery. Finally, training samples are generated based on the sample battery data, sample battery type, and fault label, that is, the battery data, battery type, and fault label are combined to form a training data set. These training samples can be used to train a neural network model so that the model learns the relationship between battery data and fault conditions, thereby achieving accurate detection of battery faults.

[0048] The training samples generated in this way can help improve the accuracy and reliability of the neural network model, thereby better realizing the real-time monitoring and prediction of battery failures. This approach helps to improve the efficiency of battery management and reduce the impact of failures on vehicle performance and safety.

[0049] As an optional embodiment, the neural network model is a random decision forest model.

[0050] Optionally, in this optional embodiment, the neural network model can be a random decision forest model. Random decision forest is an integrated learning algorithm that consists of multiple decision trees, each of which is trained independently. In a random decision forest, each decision tree will make a prediction for the input data, and then determine the final prediction result by voting or taking the average. In this embodiment, the random decision forest model will be used for battery fault detection. The training samples will determine the sample battery type and generate a data set for training by obtaining the battery data and fault conditions of the sample vehicle. The random decision forest model will use these training samples to learn the relationship between battery data and fault conditions, and finally output the detection result of whether the battery is faulty.

[0051] Compared with the neural network model, the random decision forest model has the advantages of easy interpretation and high computational efficiency. It performs well when processing large-scale data and high-dimensional features, and does not require a large amount of training data and parameter adjustment like neural networks. Therefore, using the random decision forest model for battery fault detection will improve the model's interpretability and training efficiency while still providing good prediction performance.

[0052] The training process of the random forest network is as follows: based on the original training set, the training set samples are sampled with replacement, the number of samples is m, and a total of n samples are sampled to generate n new training sets; after that, the n decision tree models are trained using the n training sets respectively; in the process of training a single decision tree model, multiple features identified need to be evaluated for feature importance, and the best feature is used for splitting. When evaluating the best feature, the Gini coefficient is used to evaluate the importance of the feature. The Gini coefficient mainly represents the impurity of the node, and its calculation formula is as follows:

[0053]

[0054] Where p i represents the proportion of samples of category i in node t, K is the number of categories, Gini(t) represents the Gini coefficient at node t, and then the change in the Gini coefficient after the node split, ΔGini(t), can be calculated using the following formula:

[0055]

[0056] In the above formula, N left 、N right is the number of samples in the left and right child nodes after the split, N t is the total number of samples at node t, Ginil(t+1) and Ginir(t+1) represent the Gini coefficients at the left and right child nodes after splitting, respectively.

[0057] Repeatedly use the above Gini coefficient calculation formula to generate decision trees, and then combine the generated decision trees to generate a random forest for classifying the battery health status, and use the different classification results of different decision trees to vote to classify the status. The training process of the random forest network mainly includes the following four steps:

[0058] ① Random sampling: Random sampling with replacement is performed from the original data set to generate multiple different training subsets, each of which contains samples of part of the original data set. These subsets are used to train each decision tree model.

[0059] ② Random selection of features: During the training process of each decision tree, a portion of features are randomly selected from all features for consideration at each node feature selection. This ensures that the features used in the training process of each decision tree are different.

[0060] ③Construct a decision tree: Use the selected feature subset to construct a decision tree for each training subset. Usually, the nodes are divided and split based on indicators such as information gain, Gini index or mean square error.

[0061] ④ Integrated prediction: When all decision trees are built, when predicting new samples, each decision tree will give its own prediction result, and the final prediction result can be determined by majority voting or average.

[0062] This enables the classification of battery health status using a random forest network.

[0063] As a specific embodiment, Figure 3 is a schematic diagram of a neural network model training process according to an optional embodiment of the present invention. Figure 3 As shown, the sample charging pile can collect data such as the SOC value, vehicle voltage, and vehicle ambient temperature of the sample vehicle. Among them, the vehicle SOC value can be classified using a density-based spatial clustering application algorithm to obtain the battery type of the sample vehicle. Among them, the density-based spatial clustering application algorithm (Density-Based Spatial Clustering of Applications with Noise, referred to as DBSCAN) is a commonly used density clustering algorithm, which is used to divide data points into different clusters according to their density. DBSCAN can aggregate data points with sufficiently high density together to form a cluster. At the same time, because noise points can be identified, DBSCAN has good robustness to outliers. This makes DBSCAN one of the commonly used clustering algorithms for processing large-scale data and noisy data sets.

[0064] Then, the battery type, vehicle SOC value, vehicle voltage and vehicle ambient temperature can be analyzed to determine the training data flag, that is, to determine whether the battery of the vehicle is faulty. Then, the above data can be statistically analyzed to extract the mechanism characteristics and statistical characteristics to obtain training samples. Among them, the mechanism characteristics are mainly used to characterize the changes in the characteristic properties of the battery system itself during the use of the battery, while the statistical characteristics are used to characterize the usage habits of normal and abnormal batteries in the past time period. The mechanism characteristics may include: battery temperature inconsistency; battery voltage inconsistency; battery SOC-OCV curve; maximum voltage / temperature; charging power; battery capacity; temperature rise rate; charging temperature. Statistical characteristics may include: number of fast charges; charging depth; charging time; cumulative charge and discharge capacity.

[0065] The training samples can then be input into the original random forest network for training. The random forest network will output the fault detection results. When the flag bit of the detection result is 1, it indicates that the battery is normal. When the flag bit of the detection result is 0, it indicates that the battery is abnormal.

[0066] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0067] Through the description of the above implementation, those skilled in the art can clearly understand that the battery fault detection method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0068] According to an embodiment of the present invention, a battery fault detection device for implementing the above-mentioned battery fault detection method is also provided. Figure 4 is a structural block diagram of a battery fault detection device provided according to an embodiment of the present invention. Figure 4 As shown, the battery fault detection device includes: a collection module 42, a determination module 44 and a detection module 46. The battery fault detection device is described below.

[0069] The collection module 42 is used to collect battery data of the battery of the target vehicle during charging through the charging pile for charging the target vehicle.

[0070] The determination module 44 is connected to the acquisition module 42 and is used to determine the type of the battery of the target vehicle according to the battery data.

[0071] The detection module 46 is connected to the determination module 44, and is used to input the battery type and battery data into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training the original neural network model through training samples, and the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data from a sample vehicle through a sample charging pile.

[0072] It should be noted that the acquisition module 42, the determination module 44 and the detection module 46 correspond to steps S202 to S206 in the embodiment, and the examples and application scenarios implemented by the multiple modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above embodiment. It should be noted that the above modules, as part of the device, can be run in the computer terminal 10 provided in the embodiment.

[0073] An embodiment of the present invention may provide a computer device. Optionally, in this embodiment, the computer device may be located in at least one network device among multiple network devices of a computer network. The computer device includes a memory and a processor.

[0074] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the battery fault detection method and device in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, the above-mentioned battery fault detection method is realized. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0075] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: collect battery data of the battery of the target vehicle during charging through the charging pile for charging the target vehicle; determine the type of the battery of the target vehicle based on the battery data; input the battery type and battery data into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training the original neural network model through training samples, and the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data from a sample vehicle through a sample charging pile.

[0076] Optionally, the processor may also execute the program code of the following steps: inputting the battery type and battery data into a pre-trained neural network model, and the neural network model outputting the detection result of whether the battery is faulty, including: determining the historical charging characteristics through the historical data recorded by the charging pile, wherein the historical charging characteristics include at least one of the following: historical fast charging times, historical charging depth, historical charging time, and cumulative charge and discharge capacity; combining the historical charging characteristics, the battery type, and the battery data, and inputting the combined data into the neural network model, and the neural network model outputting the detection result.

[0077] Optionally, the processor may also execute program code for the following steps: when the battery data includes a battery state of charge, determining the type of the battery of the target vehicle based on the battery data, including: obtaining multiple battery states of charge of the target vehicle at multiple moments respectively; combining the multiple battery states of charge with the multiple moments in a one-to-one correspondence to obtain multiple two-dimensional data points; and determining the type of the battery based on the multiple two-dimensional data points.

[0078] Optionally, the processor may also execute program code for the following steps: determining the type of battery based on multiple two-dimensional data points, including: obtaining multiple sample two-dimensional data points, wherein the types of batteries corresponding to the multiple sample two-dimensional data points have been determined, and the types of batteries corresponding to the multiple sample two-dimensional data points include lithium iron phosphate batteries and other batteries; clustering the multiple sample two-dimensional data points and the multiple two-dimensional data points to obtain multiple classifications; and determining the type of battery based on the multiple classifications.

[0079] Optionally, the processor may also execute the program code of the following steps: the training samples are determined in the following manner: based on the sample charging pile, the sample battery data of the sample vehicle is obtained; the sample battery data is classified to determine the sample battery type; the fault condition of the sample vehicle is obtained to determine the fault label corresponding to the sample battery data; and the training samples are generated according to the sample battery data, the sample battery type and the fault label.

[0080] Optionally, the processor may also execute program code of the following steps: the neural network model is a random decision forest model.

[0081] By adopting the embodiment of the present invention, a solution for battery fault detection is provided. The battery data of the battery of the target vehicle when charging is collected through the charging pile for charging the target vehicle; the type of the battery of the target vehicle is determined according to the battery data; the type of the battery and the battery data are input into a pre-trained neural network model, and the neural network model outputs the detection result of whether the battery is faulty, wherein the neural network model is obtained by training the original neural network model through training samples, and the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data from the sample vehicle through the sample charging pile, so as to achieve the purpose of performing battery fault detection at the charging end according to the relevant data of the charging vehicle, thereby achieving the technical effect of detecting battery faults when charging the electric vehicle at the charging end, and further solving the technical problem that the battery fault detection method in the related technology cannot be applied to the charging end.

[0082] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a non-volatile storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0083] The embodiment of the present invention further provides a non-volatile storage medium. Optionally, in this embodiment, the non-volatile storage medium can be used to store the program code executed by the battery fault detection method provided in the above embodiment.

[0084] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0085] Optionally, in this embodiment, the non-volatile storage medium is configured to store program codes for executing the following steps: collecting battery data of the battery of the target vehicle during charging through a charging pile that charges the target vehicle; determining the type of the battery of the target vehicle based on the battery data; inputting the battery type and battery data into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training an original neural network model through training samples, the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data on a sample vehicle through a sample charging pile.

[0086] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: inputting the battery type and battery data into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, including: determining historical charging characteristics through historical data recorded by the charging pile, wherein the historical charging characteristics include at least one of the following: historical fast charging times, historical charging depth, historical charging time, and cumulative charge and discharge capacity; combining the historical charging characteristics, the battery type, and the battery data, and inputting the combined data into the neural network model, and the neural network model outputs the detection result.

[0087] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: when the battery data includes a battery state of charge, determining the type of battery of the target vehicle based on the battery data, including: obtaining multiple battery states of charge of the target vehicle at multiple moments respectively; combining the multiple battery states of charge with the multiple moments in a one-to-one correspondence to obtain multiple two-dimensional data points; and determining the type of battery based on the multiple two-dimensional data points.

[0088] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: determining the type of battery based on multiple two-dimensional data points, including: obtaining multiple sample two-dimensional data points, wherein the types of batteries corresponding to the multiple sample two-dimensional data points have been determined, and the types of batteries corresponding to the multiple sample two-dimensional data points include lithium iron phosphate batteries and other batteries; clustering the multiple sample two-dimensional data points and the multiple two-dimensional data points to obtain multiple classifications; determining the type of battery based on the multiple classifications.

[0089] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for performing the following steps: the training sample is determined in the following manner: based on the sample charging pile, the sample battery data of the sample vehicle is obtained; the sample battery data is classified to determine the sample battery type; the fault condition of the sample vehicle is obtained to determine the fault label corresponding to the sample battery data; and the training sample is generated according to the sample battery data, the sample battery type and the fault label.

[0090] Optionally, in this embodiment, the non-volatile storage medium is configured to store program codes for executing the following steps: the neural network model is a random decision forest model.

[0091] An embodiment of the present invention further provides a computer program product, including a computer program, which implements the steps of the battery fault detection method in each embodiment of the present application when the computer program is executed by a processor.

[0092] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0093] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0094] 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.

[0095] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0096] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0097] 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 non-volatile storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0098] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A battery failure detection method, characterized in that: include: Collecting battery data of the target vehicle's battery during charging through a charging pile for the target vehicle; Determining the type of the battery of the target vehicle according to the battery data; The battery type and the battery data are input into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training an original neural network model through training samples, and the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data from a sample vehicle through a sample charging pile.

2. The method according to claim 1, characterized in that The step of inputting the type of the battery and the battery data into a pre-trained neural network model, and outputting a detection result of whether the battery is faulty by the neural network model, comprises: Determine historical charging characteristics through historical data recorded by the charging pile, wherein the historical charging characteristics include at least one of the following: historical fast charging times, historical charging depth, historical charging duration, and cumulative charging and discharging capacity; The historical charging characteristics, the type of the battery and the battery data are combined, and the combined data are input into the neural network model, and the neural network model outputs the detection result.

3. The method according to claim 1, characterized in that In a case where the battery data includes a battery state of charge, determining the type of the battery of the target vehicle according to the battery data includes: Respectively obtaining multiple battery charge states of the target vehicle at multiple times; Combine the multiple battery charge states and the multiple time points in a one-to-one correspondence to obtain multiple two-dimensional data points; The type of the battery is determined based on the plurality of two-dimensional data points.

4. The method according to claim 3, characterized in that Determining the type of the battery according to the plurality of two-dimensional data points includes: Acquire a plurality of sample two-dimensional data points, wherein the types of batteries corresponding to the plurality of sample two-dimensional data points have been determined, and the types of batteries corresponding to the plurality of sample two-dimensional data points include lithium iron phosphate batteries and other batteries; Clustering the plurality of sample two-dimensional data points and the plurality of two-dimensional data points to obtain a plurality of classifications; Based on the multiple classifications, the type of the battery is determined.

5. The method according to claim 1, characterized in that The training samples are determined in the following manner: Based on the sample charging pile, obtaining sample battery data of the sample vehicle; Classifying the sample battery data to determine the sample battery type; Obtaining the fault condition of the sample vehicle and determining the fault label corresponding to the sample battery data; The training sample is generated according to the sample battery data, the sample battery type and the fault label.

6. The method according to any one of claims 1 to 5, characterized in that The neural network model is a random decision forest model.

7. A battery fault detection device, characterized in that: include: A collection module, used to collect battery data of the battery of the target vehicle during charging through a charging pile for charging the target vehicle; A determination module, used to determine the type of the battery of the target vehicle according to the battery data; A detection module is used to input the type of battery and the battery data into a pre-trained neural network model, and the neural network model outputs a detection result of whether the battery is faulty, wherein the neural network model is obtained by training an original neural network model through training samples, and the training samples include sample battery data, sample battery type and fault label, and the sample battery data is obtained by collecting data from a sample vehicle through a sample charging pile.

8. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored program, wherein when the program is executed, the device where the non-volatile storage medium is located is controlled to execute the battery fault detection method according to any one of claims 1 to 6.

9. A computer device, characterized in that: include: Memory and processor, The memory stores a computer program; The processor is used to execute the computer program stored in the memory, and when the computer program is run, the processor executes the battery fault detection method according to any one of claims 1 to 6.

10. A computer program product comprising computer instructions, characterized in that: The computer instructions are executed by a processor to implement the battery failure detection method according to any one of claims 1 to 6.

Images