Battery voltage detection method, device, equipment, medium and product

Abstract

The invention discloses a battery voltage detection method, device and equipment, a medium and a product. The method comprises the following steps: acquiring sample voltage data, the sample voltage data comprising a sample cell voltage, a sample battery voltage and a sample battery state, and the sample voltage data corresponding to a sample cell failure mode in the sample battery state; inputting the sample cell voltage, the sample battery state and the sample cell failure mode into a first model, and predicting to obtain a predicted battery voltage, the first model being a to-be-trained machine learning model; based on the difference between the predicted battery voltage and the sample battery voltage, iteratively training the first model to obtain a second model, the second model being used for performing cell failure detection based on the battery operation data of the target battery system to obtain cell failure early warning information, the battery cell failure early warning information is used for indicating a first battery cell failure mode existing in the target battery system under the battery operation data. The method can improve the cell fault detection efficiency of the battery pack.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of battery detection, and in particular relates to a battery voltage detection method, device, equipment, medium and product. BACKGROUND

[0002] Fault prediction of an energy storage battery pack is one of the key links to ensure safe operation of a battery system.

[0003] In the related art, since an energy storage battery pack is usually obtained by combining multiple battery cells, when performing fault prediction, it is usually necessary to collect diversified information such as battery cell voltages, battery voltages, operating environments of the battery pack, and the like corresponding to each battery cell, and comprehensively analyze the above information to achieve fault prediction, so the fault prediction efficiency is low. SUMMARY

[0004] The present application provides a battery voltage detection method, device, equipment, medium and product, and the technical scheme provided by the present application includes the following aspects.

[0005] According to an aspect of the present application, a battery voltage detection method is provided, which includes:

[0006] Obtaining sample voltage data, the sample voltage data including sample battery cell voltages, sample battery voltages and sample battery states, the sample battery voltage being a voltage of a battery pack formed by at least one battery cell, the sample voltage data corresponding to a sample battery cell failure mode under the sample battery state, the sample battery cell failure mode being used to indicate a reason for abnormal working state of a battery cell of a battery;

[0007] Inputting the sample battery cell voltages, the sample battery states and the sample battery cell failure mode into a first model to obtain a predicted battery voltage, the first model being a machine learning model to be trained;

[0008] Iteratively training the first model based on a difference between the predicted battery voltage and the sample battery voltage to obtain a second model;

[0009] The second model is used to perform battery cell failure detection based on battery operating data of a target battery system, obtain battery cell failure early warning information, the battery operating data being used to indicate an operating state of a battery pack in the target battery system, and the battery cell failure early warning information being used to indicate a first battery cell failure mode existing in the target battery system under the battery operating data.

[0010] According to an aspect of the present application, a battery voltage detection device is provided, which includes:

[0011] an acquisition module, configured to acquire sample voltage data, the sample voltage data including a sample cell voltage, a sample battery voltage, and a sample battery state, the sample battery voltage being the voltage of a battery pack formed by at least one cell, the sample voltage data corresponding to a sample cell failure mode under the sample battery state, the sample cell failure mode being used to indicate a cause of an abnormal working state of a battery cell;

[0012] a training module, configured to input the sample cell voltage, the sample battery state, and the sample cell failure mode into a first model to predict a predicted battery voltage, where the first model is a machine learning model to be trained;

[0013] The training module is further configured to iteratively train the first model based on the difference between the predicted battery voltage and the sample battery voltage to obtain a second model;

[0014] Among them, the second model is used to perform cell failure detection based on the battery operation data of the target battery system and obtain cell failure warning information, the battery operation data is used to indicate the operating status of the battery pack in the target battery system, and the cell failure warning information is used to indicate the first cell failure mode existing in the target battery system under the battery operation data.

[0015] In some optional embodiments, the acquisition module further includes:

[0016] A construction unit is used to establish a cell equivalent model, wherein the cell equivalent model is a model that simulates the electrochemical behavior of the cell;

[0017] A generating unit is configured to generate the sample voltage data using the cell equivalent model.

[0018] In some optional embodiments, the generating unit is further used to input a second cell failure mode into the cell equivalent model, simulate the cell failure condition corresponding to the second cell failure mode through the cell equivalent model, and obtain the cell voltage, battery voltage and battery status under the cell failure condition as the sample voltage data.

[0019] In some optional embodiments, the battery cell equivalent model is used to simulate a battery pack obtained by connecting multiple battery cells in series;

[0020] The construction unit is further configured to group the multiple battery cells in the battery cell equivalent model to obtain at least one battery cell combination;

[0021] The construction unit is further configured to generate an extended equivalent model based on the at least one battery cell combination;

[0022] The generating unit is further configured to input the second cell failure mode into the extended equivalent model, simulate the cell failure condition corresponding to the second cell failure mode through the extended equivalent model, and obtain the cell combination voltage, the battery voltage, and the battery status of the cell combination under the cell failure condition as the sample voltage data.

[0023] In some optional embodiments, the apparatus further includes: a prediction module, the prediction module including:

[0024] a generating unit, configured to establish a voltage database using the second model, the voltage database including a plurality of candidate battery voltages, and candidate battery states and a third battery cell failure mode corresponding to the candidate battery voltages, the third battery cell failure mode being used to determine the first battery cell failure model indicated by the battery cell failure warning information by comparing the candidate battery voltages with the candidate battery states;

[0025] A comparison unit is used to generate the cell failure warning information based on a comparison result of the battery operation data in the battery database.

[0026] In some optional embodiments, the battery operating data includes a battery operating voltage and a battery operating status of the target battery system, where the battery operating status is used to indicate an operating status of a battery pack in the target battery system;

[0027] The comparison unit is further configured to determine a target battery voltage that matches the battery operating voltage from the multiple candidate battery voltages, the target battery voltage corresponding to at least one candidate battery state in the battery database;

[0028] The comparison unit is further configured to obtain, from the voltage database, the first cell failure mode corresponding to the target battery voltage under the battery operating state;

[0029] The comparison unit is further configured to generate battery cell failure warning information corresponding to the first battery cell failure mode.

[0030] In some optional embodiments, the acquisition module is further configured to acquire an end-to-end target battery voltage of a target battery pack in the target battery system; and acquire a battery operating status of a battery pack in the target battery system;

[0031] The acquisition module is further configured to use the target battery voltage and the battery operating status as the battery operating data.

[0032] According to an aspect of the embodiments of the present application, a terminal device is provided, which comprises a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the battery voltage detection method.

[0033] According to an aspect of the embodiments of the present application, a computer readable storage medium is provided, which stores a computer program, the computer program being loaded and executed by a processor to implement the battery voltage detection method.

[0034] According to an aspect of the embodiments of the present application, a computer program product is provided, which comprises a computer program stored in a computer readable storage medium, the computer program being read and executed by a processor from the computer readable storage medium to implement the battery voltage detection method.

[0035] The technical solutions provided by the embodiments of the present application can bring the following beneficial effects:

[0036] The second model capable of predicting the voltage data of the battery pack is obtained by training the sample voltage data, and the cell failure detection is performed on the current battery operation data by the second model, so as to prewarn the possible battery failure mode of the battery pack. That is, since the second model obtained by training can perceive the mutual relationship between the cell voltage, the battery voltage, the battery state and the cell failure mode, the second model can quickly realize the cell failure detection according to the current operation state of the battery pack, thereby improving the cell fault detection efficiency of the battery pack. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0038] Figure 1 is a schematic diagram of a computer system provided by an embodiment of the present application;

[0039] Figure 2 is a flowchart of a battery voltage detection method provided by an embodiment of the present application;

[0040] Figure 3 is a schematic diagram of a cell equivalent model provided by an embodiment of the present application;

[0041] Figure 4 is a flowchart of a battery voltage detection method provided by an embodiment of the present application;

[0042] Figure 5 is a flowchart of a battery voltage detection method provided by an embodiment of the present application;

[0043] Figure 6 is a flowchart of a battery voltage detection method provided by an embodiment of the present application;

[0044] Figure 7 is a block diagram of a battery voltage detection device provided by an embodiment of the present application;

[0045] Figure 8 is a block diagram of a battery voltage detection device provided by an embodiment of the present application;

[0046] Figure 9 is a structural block diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0047] To make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the drawings.

[0048] First, the terms involved in the embodiments of the present application are briefly introduced.

[0049] Battery cell: the most basic energy storage unit in a battery system, usually refers to a single battery unit, which is composed of positive electrode, negative electrode, electrolyte and separator, etc. The battery cell stores and releases energy through chemical reaction, and its performance directly affects the efficiency and life of the entire battery system. Battery cells can be various types of chemical batteries, such as lithium-ion batteries, nickel-hydrogen batteries, etc., which are widely used in portable electronic devices, electric vehicles and energy storage systems.

[0050] Battery pack: a larger energy storage system composed of multiple battery cells connected in series, parallel or mixed. The battery pack can provide higher voltage or capacity than a single battery cell to meet the needs of different devices. In the battery pack, the battery management system (BMS) plays a crucial role, which is responsible for monitoring the voltage, current, temperature and other parameters of the battery cell, ensuring the safe and efficient operation of the battery pack, and optimizing the service life of the battery. Battery packs are widely used in mobile devices, home energy storage and grid energy storage, etc.

[0051] Please refer to Figure 1 , which shows a schematic diagram of a computer system provided by an embodiment of the present application, which is used to implement the battery voltage detection method provided by the embodiments of the present application. The computer system includes: a battery monitoring device 110, a target battery system 120.

[0052] The battery monitoring device 110 is installed and runs an application program with a fault prediction function for the target battery system 120.

[0053] In some embodiments, the battery monitoring device 110 can be implemented as a terminal device, and the device type of the terminal device includes at least one of a smartphone, a notebook computer, a desktop computer, a tablet computer, a smart television, a smart robot, an industrial smart device, and the like.

[0054] In other embodiments, the battery monitoring device 110 can be implemented as a server, which can be implemented as a physical server, a cloud server in the cloud, or a node in a blockchain system. The server includes at least one of a single server, multiple servers, a cloud computing platform, and a virtualization center.

[0055] In other embodiments, the battery monitoring device 110 can be implemented as a combination of a terminal device and a server. Optionally, the server undertakes the main computing work, and the terminal device undertakes the secondary computing work; or the server undertakes the secondary computing work, and the terminal device undertakes the main computing work; or the server and the terminal device adopt a distributed computing architecture for collaborative computing.

[0056] The target battery system 120 is a battery system including at least one battery pack, and the battery pack is a battery composed of at least one battery cell. In some embodiments, the target battery system 120 further includes a BMS for monitoring and managing the state of each battery pack, including voltage, current, temperature, state of charge (SOC), and the like, to ensure the safety and optimal performance of the battery pack.

[0057] In some embodiments, the target battery system 120 corresponds to at least one data output interface, and the battery monitoring device 110 obtains battery operation data from the target battery system 120 through the at least one data output interface.

[0058] The battery monitoring device 110 obtains sample voltage data, the sample voltage data includes sample cell voltages, sample battery voltages, and sample battery states, the sample battery voltage is the voltage of a battery group formed by at least one cell, the sample voltage data corresponds to a sample cell failure mode under a sample battery state, and the sample cell failure mode is used to indicate the cause of the abnormal working state of the cell of the battery.

[0059] In one example, the battery monitoring device 110 is implemented as a combination of a terminal device and a server. The server obtains sample voltage data, the sample voltage data includes sample cell voltages, sample battery voltages, and sample battery states, the sample battery voltage is the voltage of a battery group formed by at least one cell, the sample voltage data corresponds to a sample cell failure mode under a sample battery state, and the sample cell failure mode is used to indicate the cause of the abnormal working state of the cell of the battery. The sample cell voltages, the sample battery states, and the sample cell failure mode are input into a first model to obtain a predicted battery voltage, the first model is a machine learning model to be trained. The first model is iteratively trained based on the difference between the predicted battery voltage and the sample battery voltage to obtain a second model, the second model is used to predict the voltage data output of the battery group. The server sends the trained second model to the terminal device, and the second model is deployed on the terminal device. The second model is used to detect cell failure based on battery operation data to obtain cell failure warning information, and the cell failure warning information is used to indicate a first cell failure mode existing in the battery operation data.

[0060] Please refer to Figure 2 , which shows a flowchart of a battery voltage detection method provided by an embodiment of the application. In this method, the battery monitoring device 110 in Figure 1 executes the method. The method can include at least one of the following steps (210-230).

[0061] Step 210: Obtain sample voltage data.

[0062] The sample voltage data includes sample cell voltages, sample battery voltages, and a sample battery state. The sample battery voltage is the voltage of a battery group formed by at least one cell. The sample voltage data corresponds to a sample cell failure mode in the sample battery state. The sample cell failure mode is used to indicate the cause of the abnormal working state of the cell of the battery.

[0063] The sample battery state is used to indicate the running state of the battery group. Optionally, the sample battery state includes at least one state information of a cell temperature, a battery group temperature, a cell SOC, and a battery group SOC.

[0064] In some embodiments, the sample cell failure mode is used to indicate the cell failure mode of the sample voltage data and the sample battery state in a specified cell failure working condition.

[0065] Optionally, the cell failure mode includes a safety failure mode and a non-safety failure mode.

[0066] The safety failure mode includes:

[0067] · Internal positive and negative electrode short circuit of the cell

[0068] The cause of the cell failure mode includes defects in the production process of the cell, such as poor alignment of the pole piece, damage to the diaphragm, and the like; and / or, long-term vibration, external force impact, and the like, which causes the cell to deform and makes the positive and negative electrodes contact. The mode influence of the cell failure mode includes that a large amount of heat is generated inside the battery, which causes the temperature to rise sharply, and may cause smoking, fire, and even explosion, which is difficult to prevent by external safety devices.

[0069] · Battery cell liquid leakage

[0070] The cause of the cell failure mode includes damage by external force, collision, non-standard installation, and the like, which causes the sealing structure to be damaged; and / or, welding defects, insufficient amount of sealing glue, and the like in the manufacturing process. The mode influence of the cell failure mode includes that the insulation failure of the battery pack is caused, and if multiple points are insulated, an external short circuit may be caused, thereby increasing the risk of fire.

[0071] · Cell swelling

[0072] The cause of the cell failure mode includes that a side reaction occurs inside the battery to generate gas, such as the reaction of impurities and moisture in the electrolyte to generate gas; and / or, under the abuse conditions of overcharging, overdischarging, high temperature, and the like, the electrolyte is decomposed to generate gas. The mode influence of the cell failure mode includes that the swelling may cause damage to the battery structure, and further cause problems such as liquid leakage and short circuit.

[0073] The non-safety failure mode includes:

[0074] · Poor capacity consistency

[0075] The causes of this cell failure mode include insufficient battery production and manufacturing process technology, such as uneven coating of the pole piece, long storage time of the battery, temperature difference and inconsistent current during charging and discharging, etc. The mode impact of this cell failure mode includes affecting the overall performance and service life of the battery pack, reducing the reliability and efficiency of the battery system.

[0076] • Excessive self-discharge

[0077] The causes of this cell failure mode include incomplete internal chemical reaction of the battery, or the presence of micro-short circuit, etc. The mode impact of this cell failure mode includes causing rapid loss of battery power, shortening the standby time and service life of the battery.

[0078] • Thermal runaway

[0079] The causes of this cell failure mode include the generation of exothermic reaction of the battery pack under high temperature conditions, which causes the temperature of the battery to rise rapidly. The mode impact of this cell failure mode includes causing the battery to burn or explode.

[0080] • Low-temperature discharge capacity reduction

[0081] The causes of this cell failure mode include the decline of ion conduction ability of the electrolyte and the slowing down of the internal electrochemical reaction rate of the battery under low temperature conditions. The mode impact of this cell failure mode includes a significant reduction in the discharge capacity of the battery in a low temperature environment, affecting the normal use of the device in a low temperature environment.

[0082] • Battery capacity attenuation

[0083] The causes of this cell failure mode include the gradual reduction of the activity of the positive and negative materials, the decomposition of the electrolyte, and the internal side reactions of the battery during use. The mode impact of this cell failure mode includes the gradual decline of the capacity of the battery and the shortening of the service life.

[0084] Among them, the cell failure condition refers to the running condition or running environment of the cell in the battery pack that produces abnormal operation state, and the cell failure condition includes system temperature, system humidity, system pressure, system vibration and other information indicating the running condition or running environment.

[0085] That is, the cell failure condition is used to indicate at least one external condition of the cell running in the battery pack, and the at least one external condition will affect the internal chemical reaction and physical state of the cell, and the cell failure mode is the specific failure form of the cell after running under the cell failure condition for a certain period of time. Alternatively, different cell failure conditions may lead to different cell failure modes, or under the same cell failure condition, the cell may exhibit multiple cell failure modes.

[0086] Optionally, the cell failure condition includes a high-temperature condition, an overcharge condition, an overdischarge condition, a high-humidity condition, a vibration condition, etc.

[0087] In some embodiments, when the battery pack is composed of a plurality of cells, the relationship between the sample battery voltage and the sample cell voltage corresponding to the battery pack depends on the connection mode of the battery pack, and optionally, the connection mode of the cells in the battery pack includes series connection and / or parallel connection; when the connection mode of the cells in the battery pack is series connection, the sample battery voltage is equal to the sum of the sample cell voltages of all the cells in the battery pack; when the connection mode of the cells in the battery pack is parallel connection, the sample battery voltage is equal to a single sample cell voltage in the battery pack, and the sample cell voltages between all the cells in parallel are the same.

[0088] Optionally, the sample voltage data can be obtained in at least one of the following ways:

[0089] Firstly, historical voltage data of the battery system in a historical running period is obtained as sample voltage data.

[0090] Optionally, the battery system described above includes a target battery system, which is a battery system that needs to be detected for battery voltage in the embodiments of the present application. Optionally, the battery system described above can also include other battery systems, which are battery systems deployed in other scenarios.

[0091] Illustratively, a running log of the battery system in a historical running period is obtained, and the running log includes historical cell voltages and historical battery voltages detected when a cell failure mode occurs in the historical running period, i.e., the historical cell voltages are taken as sample cell voltages, the historical battery voltages are taken as sample battery voltages, and the cell failure mode that occurs is labeled as a sample cell failure mode corresponding to the sample cell voltages and the sample battery voltages.

[0092] Secondly, sample voltage data is generated through a cell equivalent model.

[0093] The cell equivalent model is a model simulating the electrochemical behavior of the cell. Optionally, the cell equivalent model can be implemented as a mathematical model or as an analog circuit model, wherein when the cell equivalent model is implemented as a mathematical model, the cell equivalent model is a mathematical equation describing the relationship between the voltage, current, temperature, state of charge, etc. of the cell; and when the cell equivalent model is implemented as an analog circuit model, the cell equivalent model is a virtual circuit built by virtual electronic components.

[0094] Illustratively, a cell equivalent model is established, and sample voltage data is generated through the cell equivalent model.

[0095] Optionally, the cell equivalent model comprises at least one of a circuit equivalent model, an electrochemical model, and a hybrid model. The circuit equivalent model is a mathematical model that simulates internal processes of the cell by circuit elements. Optionally, the circuit equivalent model comprises a Thevenin equivalent circuit model, a PNGV (Partnership for a New Generation of Vehicles) model, a Rint model, etc. The electrochemical model comprises a P2D (Pseudo-Two-Dimensional) model, a single particle model (SPM), a distributed parameter model (DPM), etc., which are not limited herein.

[0096] In some embodiments, the corresponding cell equivalent model is constructed according to a combination of cells in the battery pack in the target battery system.

[0097] In some embodiments, the second cell failure mode is input to the cell equivalent model, a cell failure condition corresponding to the second cell failure mode is simulated by the cell equivalent model, and a cell voltage, a battery voltage, and a battery state under the cell failure condition are obtained as sample voltage data. That is, the cell equivalent model outputs, according to the indicated third cell failure mode, a cell voltage and a battery state exhibited by a cell of the battery pack under a cell failure condition corresponding to the third cell failure mode, and a battery voltage of the battery pack.

[0098] In some embodiments, the third cell failure mode comprises at least one of an internal positive and negative electrode short circuit of the cell, a battery monomer liquid leakage, a cell swelling and expansion capacity consistency difference, an excessive self-discharge, a thermal runaway, a low-temperature discharge capacity reduction, a battery capacity attenuation, etc.

[0099] In some embodiments, the combination manner of the battery cells in the battery cell equivalent model is reconstructed from the battery cell length dimension to obtain an extended equivalent model corresponding to at least one battery cell combination manner; and sample voltage data is generated by using the extended equivalent model. For example, the battery cell equivalent model is used to simulate a battery pack obtained by connecting a plurality of battery cells in series, the plurality of battery cells in the battery cell equivalent model are grouped to obtain at least one battery cell combination, an extended equivalent model is generated based on the at least one battery cell combination, a second battery cell failure mode is input to the extended equivalent model, and the extended equivalent model is used to simulate a battery cell failure condition corresponding to the second battery cell failure mode to obtain a battery cell combination voltage, a battery voltage and a battery state of the battery cell combination under the battery cell failure condition as sample voltage data. That is, the sample voltage data generated by using the extended equivalent model includes the battery cell combination voltage, the battery voltage and the battery state, wherein the battery cell combination voltage is the voltage of the battery cell combination obtained by grouping the plurality of battery cells, and in this case, the battery cell combination is regarded as a whole to output the battery cell voltage (battery cell combination voltage), so that more battery operating states are simulated by using the extended equivalent model.

[0100] Specifically, generating the extended equivalent model based on the at least one battery cell combination is implemented by: regarding the at least one battery cell combination as a whole battery cell respectively, and generating an updated virtual circuit as the extended equivalent model, wherein the extended equivalent model outputs a battery cell combination voltage as sample voltage data when the model battery operating state is simulated, and the battery cell combination is used as a voltage output unit of the battery cell voltage.

[0101] In some embodiments, reconstructing the combination manner of the battery cells in the battery cell equivalent model from the battery cell length dimension includes regarding a target number of battery cells in the battery cell equivalent model as a combined battery cell to recombine the plurality of battery cells in the battery cell equivalent model to obtain the extended equivalent model.

[0102] For example, as shown in FIG. 3, the battery cell equivalent model 301 is an equivalent circuit composed of 96 simulation battery cells, the extended equivalent model 302 is obtained by regarding two battery cells as a combined battery cell, and the extended equivalent model 303 is obtained by regarding three battery cells as a combined battery cell. Figure 3

[0103] In some embodiments, the sample battery cell voltage, the sample battery state and the sample battery cell failure mode are input to the first model to obtain a predicted battery voltage.

[0104] In some embodiments, the first model is a machine learning model to be trained. In the embodiments of the present application, the first model is used to implement a prediction task of the battery voltage of the corresponding battery pack by using the single battery cell voltage and the battery cell failure mode.

[0105] ​Optionally, the first model can be implemented by a neural network model such as a Convolutional Neural Network (CNN), a Feedforward Neural Network (FNN), a Residual Network (ResNet), a Transformer, etc., which is not specifically limited herein. In an example, the first model is implemented by a regression model to predict the battery voltage.

[0106] In some embodiments, the sample voltage data includes a plurality of time-sequentially arranged sample cell voltages, sample battery states, and sample battery voltages, where the plurality of time-sequentially arranged sample cell voltages form a sample cell voltage sequence, the plurality of time-sequentially arranged sample battery states form a sample battery state sequence, and the plurality of time-sequentially arranged sample battery voltages form a sample battery voltage sequence, and the time stamps of the corresponding sequence items of the cell voltage sequence, the sample battery state sequence, and the sample battery voltage sequence are the same.

[0107] Illustratively, the sample cell voltage sequence, the sample battery state sequence, and the sample failure mode are input into the first model, and a predicted battery voltage sequence corresponding to the predicted sample cell voltage sequence is predicted by the first model.

[0108] In some embodiments, a sliding window of a preset length is used to extract a sample cell voltage from the sample cell voltage, and the sample cell voltage, the sample battery state corresponding to the sample cell voltage, and the sample failure mode are input into the first model, and a predicted battery voltage sequence is predicted by the first model.

[0109] At step 230, the first model is iteratively trained based on the difference between the predicted battery voltage and the sample battery voltage to obtain a second model.

[0110] In some embodiments, the predicted battery voltage and the sample battery voltage are input into a preset loss function to obtain a loss value, and the model parameters of the first model are updated using a backpropagation algorithm based on the loss value to obtain the second model.

[0111] Optionally, the preset loss function can be implemented as at least one of a Cross-Entropy Loss function, a Mean Squared Error Loss (MSE) function, a log loss function, a Least Absolute Deviations Loss (L1 Loss) function, etc., which is not specifically limited herein.

[0112] In some embodiments, when the input data of the first model is the sample cell voltage sequence and the sample failure model, the output predicted battery voltage sequence and the sample battery voltage sequence in the sample voltage data are input into a preset loss function to obtain a loss value, and the model parameters of the first model are updated using a back propagation algorithm on the basis of the loss value to obtain a second model.

[0113] In the embodiments of the present application, the second model is used to predict the voltage data output of the battery pack for cell failure detection.

[0114] In some embodiments, please refer to Figure 4 which shows a flowchart of a battery voltage detection method according to an example embodiment of the present application. The method can include at least one of the following steps (240-250).

[0115] In step 240, battery operation data of a target battery system is obtained.

[0116] In the embodiments of the present application, the target battery system is a battery system that needs to be detected for battery voltage, and the battery operation data is used to indicate the operation state of the battery pack in the target battery system.

[0117] In some embodiments, the target battery system includes a target battery pack composed of target cells. Optionally, the target battery pack includes at least one target cell. Optionally, when the target battery pack includes multiple target cells, the connection mode between the multiple target cells includes series connection and / or parallel connection.

[0118] In some embodiments, the battery operation data includes the target battery voltage corresponding to the target battery pack in the target battery system, and the battery operation state of the target battery pack in the target battery system. Illustratively, the target battery voltage from the end to end of the target battery pack in the target battery system is collected; and the system environment information in the target battery system is collected to obtain the battery operation state of the target battery pack; and the target battery voltage and the battery operation state are taken as the battery operation data.

[0119] Optionally, the battery operation state includes at least one of the state information of the cell temperature, the battery pack temperature, the cell SOC and the battery pack SOC.

[0120] In some embodiments, the target battery voltage and the battery operation state are output by the BMS in the target battery system.

[0121] In some embodiments, the detection mode of the target battery voltage from the end to end of the target battery pack is implemented by detecting the cell voltage corresponding to each cell in the target battery pack, and calculating the target battery voltage corresponding to the target battery pack according to the connection mode between the cells in the target battery pack.

[0122] In some embodiments, when the plurality of battery cells in the target battery pack are connected in parallel, the detection of the target battery voltage of the target battery pack can be implemented by detecting the battery voltage of the target battery pack.

[0123] At step 250, the battery operation data is input into the second model to detect the battery cell failure, and battery cell failure warning information is obtained.

[0124] The battery cell failure warning information is used to indicate the first battery cell failure mode existing in the battery operation data.

[0125] In some embodiments, the battery operation data is input into the second model, the battery cell failure mode corresponding to the battery operation model is classified by the second model, the first battery cell failure mode is obtained, and the battery cell failure warning information is generated based on the first battery cell failure mode.

[0126] In some embodiments, the battery operation data is input into the second model, the battery cell failure mode corresponding to the battery operation model is classified by the second model, the first battery cell failure mode is obtained, and the battery cell failure warning information is generated based on the first battery cell failure mode.

[0127] In some embodiments, the battery cell failure warning information includes at least one of the first battery cell failure mode corresponding to the battery operation data, the fault handling operation corresponding to the first battery cell failure mode, and the failure cause corresponding to the first battery cell failure mode.

[0128] Optionally, different failure handling operations correspond to different cell failure modes. Illustratively, when the first cell failure mode is internal positive / negative electrode short circuit of the cell, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, cell replacement operation, and battery pack replacement operation; when the first cell failure mode is battery monomer liquid leakage, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, cell replacement operation, and battery pack replacement operation; when the first cell failure mode is poor capacity consistency of cell swelling, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, cell replacement operation, and battery pack replacement operation; when the first cell failure mode is excessive self-discharge, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, adjustment of cyclic charging strategy, adjustment of composition and structure of positive / negative electrode material of the cell, and adjustment of temperature management strategy of the battery pack; when the first cell failure mode is thermal runaway, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, starting of cooling system operation, and adjustment of combined spacing between cells in the battery pack; when the first cell failure mode is low-temperature discharge capacity reduction, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, starting of battery heating system operation, and adjustment of composition and structure of positive / negative electrode material of the cell; and when the first cell failure mode is battery capacity attenuation, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, addition of electrolyte additive operation, surface coating treatment operation performed on silicon-based negative electrode material, and adjustment of charging / discharging strategy of the battery pack.

[0129] In summary, the second model capable of predicting voltage data of the battery pack is trained by sample voltage data, and cell failure detection is performed on current battery operation data by the second model, so as to early warn possible battery failure modes of the battery pack. That is, since the trained second model can perceive the mutual relationship among cell voltage, battery voltage, and cell failure mode, the second model can quickly perform cell failure detection according to the current operation state of the battery pack, thereby improving the cell failure detection efficiency of the battery pack.

[0130] In the embodiments of the present application, when the voltage prediction realized by the machine learning model is used for cell failure early warning, the machine learning model can be further retrained and algorithm upgraded according to the result of the cell failure early warning, thereby improving the accuracy of cell failure detection.

[0131] In some optional embodiments, after the second model is trained, a voltage database is established by the second model, for battery voltage detection and failure early warning of a target battery system. Illustratively, please refer to Figure 5Fig. 2 shows a flowchart of a method for detecting battery voltage according to an example embodiment of the present application, which can include at least one of the following steps (251-252), wherein step 251 is performed after step 240.

[0132] Step 251, establishing a voltage database by using the second model.

[0133] The voltage database includes a plurality of candidate battery voltages, candidate battery states corresponding to the candidate battery voltages, and third cell failure modes, wherein the third cell failure modes are used to determine the first cell failure mode indicated by the cell failure warning information by comparing the candidate battery voltages and the candidate battery states.

[0134] In some embodiments, the voltage database is established by sampling a sampling cell voltage from a preset cell voltage range, sampling a sampling battery state from a preset battery state range, and sampling a sampling cell failure mode from a candidate cell failure mode; inputting the sampling cell voltage, the sampling battery state and the sampling cell failure mode into the second model, and outputting the corresponding battery voltage as the candidate battery voltage; and storing the above-mentioned sampling battery failure mode as the third cell failure mode and the above-mentioned rear-end battery voltage.

[0135] In some embodiments, the above-mentioned cell voltage range is a cell voltage range that can be achieved by the cell in the target battery system, and the above-mentioned battery state range is a set of battery states corresponding to the cell in the target battery system under all cell failure modes.

[0136] In some embodiments, the sampling method of the cell voltage range and the candidate cell mode can be at least one of random sampling, grouping sampling, snowball sampling, double sampling, etc., which is not limited here.

[0137] Step 252, generating cell failure warning information based on the comparison result of the battery operation data in the battery database.

[0138] In the embodiments of the present application, the target battery system is a battery system that needs to detect battery voltage, and the battery operation data is used to indicate the running state of the battery pack in the target battery system.

[0139] In some embodiments, the target battery system includes a target battery pack composed of target cells. Optionally, the target battery pack includes at least one target cell. Optionally, when the target battery pack includes a plurality of target cells, the connection mode between the plurality of target cells includes series connection and / or parallel connection.

[0140] In some embodiments, the battery operation data comprises a target battery voltage corresponding to a target battery pack in a target battery system, and a battery operation state of the target battery pack in the target battery system. Illustratively, the target battery voltage end-to-end of the target battery pack in the target battery system is collected; and the battery operation state of the battery pack in the target battery system is collected; and the target battery voltage and the battery operation state are taken as the battery operation data.

[0141] Optionally, the battery operation state comprises at least one state information of a cell temperature, a battery pack temperature, a cell SOC, and a battery pack SOC.

[0142] In some embodiments, the target battery voltage and the battery operation state are output by a BMS in the target battery system.

[0143] In some embodiments, the detection of the target battery voltage end-to-end of the target battery pack is implemented by detecting a cell voltage corresponding to each cell in the target battery pack, and calculating the target battery voltage corresponding to the target battery pack according to a connection mode between the cells in the target battery pack.

[0144] In other embodiments, when the multiple cells in the target battery pack are connected in series, the detection of the target battery voltage end-to-end of the target battery pack can also be implemented by detecting a cell combination voltage of a cell combination composed of a specified number of cells in the target battery pack, and the battery operation state of the target battery pack in the target battery system. For example, when the target battery pack comprises N cells, N / 5 is taken as the specified number to divide the N cells into 5 cell combinations, and the cell combination voltage corresponding to each cell combination is detected respectively, and the target battery voltage of the target battery pack is obtained through the cell combination voltage.

[0145] In some embodiments, when the battery operation data comprises a battery operation voltage and a battery operation state of the target battery system, the comparison between the battery operation data and the battery database is implemented by determining a target battery voltage matching the battery operation voltage from a plurality of candidate battery voltages, the target battery voltage corresponding to at least one candidate battery state in the battery database; obtaining a first cell failure mode corresponding to the target battery voltage under the battery operation state from the voltage database; and generating cell failure warning information corresponding to the first cell failure mode.

[0146] Illustratively, when the target battery voltage in the battery operation data is obtained by grouping the cells and detecting the voltage, the number of cell voltages to be detected in the cell failure detection can be reduced, thereby reducing the hardware cost of voltage acquisition; when the voltage is acquired by the BMS, the number of BMS acquisition chips can be reduced, thereby reducing the hardware cost of the BMS.

[0147] In some embodiments, the cell failure early warning information includes at least one of a first cell failure mode corresponding to the battery operation data, a failure handling operation corresponding to the first cell failure mode, and a failure cause corresponding to the first cell failure mode.

[0148] Optionally, different failure handling operations correspond to different cell failure modes. Illustratively, when the first cell failure mode is internal positive / negative electrode short circuit of the cell, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, cell replacement operation, and battery pack replacement operation; when the first cell failure mode is battery monomer liquid leakage, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, cell replacement operation, and battery pack replacement operation; when the first cell failure mode is cell swelling and poor capacity consistency, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, cell replacement operation, and battery pack replacement operation; when the first cell failure mode is excessive self-discharge, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, adjusting the cycle charging strategy, adjusting the composition and structure of the positive / negative electrode material of the cell, and adjusting the temperature management strategy of the battery pack; when the first cell failure mode is thermal runaway, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, starting the cooling system operation, and adjusting the combined spacing between the cells in the battery pack; when the first cell failure mode is low-temperature discharge capacity reduction, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, battery heating system start operation, and adjusting the composition and structure of the positive / negative electrode material of the cell; and when the first cell failure mode is battery capacity attenuation, the failure handling operation indicated by the cell failure early warning information includes at least one of battery pack power-off operation, adding electrolyte additive operation, performing surface coating treatment operation on the silicon-based negative electrode material, and adjusting the charge / discharge strategy of the battery pack.

[0149] Illustratively, please refer to Figure 6The flowchart shows a battery voltage detection method provided by an example embodiment of the application, which includes a model pre-training and data generation stage 610 and an actual scene early warning stage 620, and the flow includes the following steps: 601, obtaining cell failure working condition data; 602, constructing a cell equivalent model; 603, inputting the cell failure working condition data into the cell equivalent model to generate multi-string cell voltage collection data; 604, obtaining battery operating environment data; 605, training an artificial intelligence model through the multi-string cell voltage collection data and the battery operating environment data, and obtaining a voltage prediction model through the training; 606, generating a multi-string voltage database through the voltage prediction model; 607, collecting multi-string voltage data and battery states by a BMS; 608, realizing failure early warning of a battery system through a comparison result of the multi-string voltage data and temperature information in the multi-string voltage database.

[0150] In summary, a second model capable of predicting voltage data of a battery pack is trained through sample voltage data, cell failure detection is performed on current battery operating data through the second model, and thus early warning of possible battery failure modes of the battery pack is performed. That is, since the trained second model can perceive the mutual relationship among cell voltage, battery voltage and cell failure mode, the second model can quickly realize cell failure detection according to the current operating state of the battery pack, and thus the cell fault detection efficiency of the battery pack is improved.

[0151] In the embodiment of the application, the voltage database is generated through the pre-trained second model, which can quickly establish all possible corresponding relationships between battery voltage changes and cell failure modes, and thus the generation efficiency of cell failure early warning information in the cell failure detection process is improved.

[0152] The following is a device embodiment of the application, which can be used to execute the method embodiments of the application. For details not disclosed in the device embodiments of the application, please refer to the method embodiments of the application.

[0153] Please refer to Figure 7 The block diagram shows a battery voltage detection device provided by an example embodiment of the application. The device includes:

[0154] The obtaining module 710 is configured to obtain sample voltage data, the sample voltage data including sample cell voltage, sample battery voltage and sample battery state, the sample battery voltage being the voltage of a battery pack formed by at least one cell, the sample voltage data corresponding to a sample cell failure mode under the sample battery state, and the sample cell failure mode being used to indicate the cause of abnormal working state of the cell of the battery;

[0155] The training module 720 is configured to input the sample battery cell voltage, the sample battery state, and the sample battery cell failure mode into a first model to obtain a predicted battery voltage, the first model being a machine learning model to be trained.

[0156] The training module 720 is further configured to iteratively train the first model based on a difference between the predicted battery voltage and the sample battery voltage to obtain a second model.

[0157] The second model is configured to perform cell failure detection based on battery operation data of a target battery system to obtain cell failure early warning information, the battery operation data being used to indicate an operation state of a battery pack in the target battery system, and the cell failure early warning information being used to indicate a first cell failure mode existing in the target battery system under the battery operation data.

[0158] In some optional embodiments, as shown in Figure 8 The acquisition module 710 further includes:

[0159] The construction unit 711 is configured to establish a battery cell equivalent model, the battery cell equivalent model being a model simulating electrochemical behavior of a battery cell.

[0160] The generation unit 712 is configured to generate the sample voltage data by using the battery cell equivalent model.

[0161] In some optional embodiments, the generation unit 712 is further configured to input a second cell failure mode into the battery cell equivalent model, simulate a cell failure working condition corresponding to the second cell failure mode by using the battery cell equivalent model, and obtain a battery cell voltage, a battery voltage, and a battery state under the cell failure working condition as the sample voltage data.

[0162] In some optional embodiments, the battery cell equivalent model is configured to simulate a battery pack obtained by connecting a plurality of battery cells in series.

[0163] The construction unit 711 is further configured to group the plurality of battery cells in the battery cell equivalent model to obtain at least one battery cell combination.

[0164] The construction unit 711 is further configured to generate an extended equivalent model based on the at least one battery cell combination.

[0165] The generation unit 712 is further configured to input the second cell failure mode into the extended equivalent model, simulate a cell failure working condition corresponding to the second cell failure mode by using the extended equivalent model, and obtain a battery cell combination voltage of the battery cell combination, the battery voltage, and the battery state under the cell failure working condition as the sample voltage data.

[0166] In some optional embodiments, the apparatus further includes a prediction module 730, which includes:

[0167] a generation unit 731 configured to establish a voltage database by using the second model, the voltage database including a plurality of candidate battery voltages, candidate battery states corresponding to the candidate battery voltages, and third cell failure modes corresponding to the candidate battery voltages, the third cell failure modes being used to determine the first cell failure mode indicated by the cell failure warning information by comparing the candidate battery voltages and the candidate battery states.

[0168] a comparison unit 732 configured to generate the cell failure warning information based on a comparison result of the battery operation data in the battery database.

[0169] In some optional embodiments, the battery operation data includes a battery operation voltage of the target battery system and a battery operation state, the battery operation state being used to indicate an operation state of a battery pack in the target battery system.

[0170] The comparison unit 732 is further configured to determine, from the plurality of candidate battery voltages, a target battery voltage matching the battery operation voltage, the target battery voltage corresponding to at least one candidate battery state in the battery database.

[0171] The comparison unit 732 is further configured to obtain, from the voltage database, the first cell failure mode corresponding to the target battery voltage in the battery operation state.

[0172] The comparison unit 732 is further configured to generate the cell failure warning information corresponding to the first cell failure mode.

[0173] In some optional embodiments, the acquisition module 710 is further configured to collect a target battery voltage of a target battery pack end to end in the target battery system, and collect a battery operation state of a battery pack in the target battery system.

[0174] The acquisition module 710 is further configured to use the target battery voltage and the battery operation state as the battery operation data.

[0175] It should be noted that the apparatus provided in the above embodiments, in realizing its functions, is only exemplified by the above division of functional modules, and in actual applications, the above functions can be completed by different functional modules according to needs, that is, the content structure of the device is divided into different functional modules to complete all or part of the above described functions. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is described in detail in the method embodiments, which will not be repeated here.

[0176] Reference is made to Figure 9 which shows a structural block diagram of an electronic device 900 provided by an embodiment of the present application. The electronic device 900 can be a Figure 1 The battery monitoring device 110 in the computer system shown is configured to implement the battery voltage detection method provided by the above-mentioned embodiments. Specifically:

[0177] Generally, the electronic device 900 includes a processor 901 and a memory 902.

[0178] The processor 901 can include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 901 can be implemented in the form of at least one of a DSP (Digital Signal Processing), a FPGA (Field Programmable Gate Array), and a PLA (Programmable Logic Array). The processor 901 can also include a main processor and a coprocessor. The main processor is a processor for processing data in an awake state, also known as a CPU (Central Processing Unit). The coprocessor is a low-power processor for processing data in a standby state. In some embodiments, the processor 901 can be integrated with a GPU (Graphics Processing Unit) that is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 901 can also include an AI (Artificial Intelligence) processor for processing machine learning-related computing operations.

[0179] The memory 902 can include one or more computer-readable storage media, which can be non-transitory. The memory 902 can also include a high-speed random access memory, and a non-volatile memory such as one or more disk storage devices, flash storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 902 is configured to store a computer program configured to be executed by one or more processors to implement the above-mentioned information display method.

[0180] In some embodiments, the electronic device 900 can further optionally include other components 903: a peripheral device interface and at least one peripheral device. The processor 901, the memory 902 and the peripheral device interface can be connected through a bus or a signal line. Each peripheral device can be connected with the peripheral device interface through a bus, a signal line or a circuit board. Specifically, the peripheral device includes at least one of a radio frequency circuit, a display screen, an audio circuit and a power supply.

[0181] Those skilled in the art can understand that the structure shown in the foregoing embodiments does not constitute a limitation on the electronic device 900, and the electronic device 900 can include more or fewer components than those shown in the drawings, or combine certain components, or adopt a different arrangement of components. Figure 9

[0182] In an example embodiment, a computer readable storage medium is also provided, and the storage medium stores a computer program. The computer program is executed by a processor to implement the battery voltage detection method described above. Optionally, the computer readable storage medium can include a ROM (Read-Only Memory), a RAM (Random Access Memory), a SSD (Solid State Drives) or an optical disc, etc. Among them, the random access memory can include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0183] In an example embodiment, a computer program product is also provided, and the computer program product includes a computer program stored in a computer readable storage medium. The processor of the terminal device reads the computer program from the computer readable storage medium, and the processor executes the computer program, so that the terminal device executes the battery voltage detection method described above.

[0184] It should be noted that the relevant data collection and processing in the present application should be strictly in accordance with the requirements of relevant national laws and regulations, and the informed consent or separate consent of the personal information subject should be obtained, and the subsequent data use and processing behavior should be carried out within the scope of authorization of laws and regulations and personal information subject.

[0185] ​It should be understood that "multiple" mentioned herein refers to two or more than two. The "and / or" describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone. The character " / " generally represents that the associated objects before and after it are in an "or" relationship. In addition, the step numbers described herein only exemplarily show a possible execution order between steps, and in some other embodiments, the above steps can also be executed in a non-numbered order, such as two different numbered steps being executed simultaneously, or two different numbered steps being executed in an order opposite to that shown in the figure, and the embodiments of the present application are not limited in this regard.

[0186] The above only describes exemplary embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting battery voltage, characterized in that: The method comprises: Acquiring sample voltage data, the sample voltage data including a sample cell voltage, a sample battery voltage, and a sample battery state, the sample battery voltage being the voltage of a battery pack formed by at least one battery cell, the sample voltage data corresponding to a sample cell failure mode under the sample battery state, the sample cell failure mode being used to indicate a cause of an abnormal working state of the battery cell; Inputting the sample cell voltage, the sample battery state, and the sample cell failure mode into a first model to predict a predicted battery voltage, where the first model is a machine learning model to be trained; Iteratively training the first model based on the difference between the predicted battery voltage and the sample battery voltage to obtain a second model; Among them, the second model is used to perform cell failure detection based on the battery operation data of the target battery system and obtain cell failure warning information, the battery operation data is used to indicate the operating status of the battery pack in the target battery system, and the cell failure warning information is used to indicate the first cell failure mode existing in the target battery system under the battery operation data.

2. The method according to claim 1, characterized in that The obtaining of sample voltage data includes: Establishing a cell equivalent model, wherein the cell equivalent model is a model that simulates the electrochemical behavior of the cell; The sample voltage data is generated by using the battery cell equivalent model.

3. The method according to claim 2, characterized in that The generating the sample voltage data by using the battery cell equivalent model includes: A second cell failure mode is input into the cell equivalent model, a cell failure condition corresponding to the second cell failure mode is simulated by the cell equivalent model, and the cell voltage, battery voltage and battery status under the cell failure condition are obtained as the sample voltage data.

4. The method according to claim 2, characterized in that The battery cell equivalent model is used to simulate a battery pack formed by connecting multiple battery cells in series; The method further comprises: Grouping the multiple battery cells in the battery cell equivalent model to obtain at least one battery cell combination; generating an extended equivalent model based on the at least one battery cell combination; The second battery cell failure mode is input into the extended equivalent model, and the battery cell failure condition corresponding to the second battery cell failure mode is simulated by the extended equivalent model to obtain the battery cell combination voltage, the battery voltage and the battery status of the battery cell combination under the battery cell failure condition as the sample voltage data.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Establishing a voltage database using the second model, the voltage database including a plurality of candidate battery voltages, and candidate battery states and a third battery cell failure mode corresponding to the candidate battery voltages, the third battery cell failure mode being used to determine the first battery cell failure model indicated by the battery cell failure warning information by comparing the candidate battery voltages with the candidate battery states; The cell failure warning information is generated based on a comparison result of the battery operation data in the battery database.

6. The method according to claim 5, characterized in that The battery operation data includes a battery operation voltage and a battery operation status of the target battery system, where the battery operation status is used to indicate an operation status of a battery pack in the target battery system; The generating the cell failure warning information based on the comparison result of the battery operation data in the battery database includes: determining a target battery voltage that matches the battery operating voltage from the plurality of candidate battery voltages, the target battery voltage corresponding to at least one candidate battery state in the battery database; Acquire, from the voltage database, the first cell failure mode corresponding to the target battery voltage under the battery operating state; Generate battery cell failure warning information corresponding to the first battery cell failure mode.

7. The method according to any one of claims 1 to 4, characterized in that: The obtaining of battery operation data of the target battery system includes: collecting an end-to-end target battery voltage of a target battery pack in the target battery system; and collecting a battery operating status of the battery pack in the target battery system; The target battery voltage and the battery operating state are used as the battery operating data.

8. A battery voltage detection device, characterized in that: The device comprises: an acquisition module, configured to acquire sample voltage data, the sample voltage data including a sample cell voltage, a sample battery voltage, and a sample battery state, the sample battery voltage being the voltage of a battery pack formed by at least one cell, the sample voltage data corresponding to a sample cell failure mode under the sample battery state, the sample cell failure mode being used to indicate a cause of an abnormal working state of a battery cell; a training module, configured to input the sample cell voltage, the sample battery state, and the sample cell failure mode into a first model to predict a predicted battery voltage, where the first model is a machine learning model to be trained; The training module is further configured to iteratively train the first model based on the difference between the predicted battery voltage and the sample battery voltage to obtain a second model; Among them, the second model is used to perform cell failure detection based on the battery operation data of the target battery system and obtain cell failure warning information, the battery operation data is used to indicate the operating status of the battery pack in the target battery system, and the cell failure warning information is used to indicate the first cell failure mode existing in the target battery system under the battery operation data.

9. A terminal device, characterized in that: The terminal device includes a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the battery voltage detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which is loaded and executed by a processor to implement the battery voltage detection method according to any one of claims 1 to 7.

11. A computer program product, characterized in that The computer program product includes a computer program, which is stored in a computer-readable storage medium. A processor reads and executes the computer program from the computer-readable storage medium to implement the battery voltage detection method according to any one of claims 1 to 7.

Images