A battery self-discharge abnormality monitoring and early warning method, system, device and medium

Abstract

The application provides a battery self-discharge abnormality monitoring and early warning method, system, device and medium, the method comprising: collecting running state data of a vehicle battery; inputting the running state data into a battery self-discharge early warning model to obtain battery state prediction information, wherein the battery self-discharge early warning model is obtained by model pre-training based on static data and dynamic data, the static data is a corresponding relationship between an end voltage sudden drop value and a self-discharge equivalent resistance value of the battery in a static state, and the dynamic data is a difference between battery parameters of a normal battery and a self-discharge fault battery in a use state; and matching a corresponding abnormality processing strategy according to the battery state prediction information. The application can monitor self-discharge of the battery throughout the whole life cycle, thereby guaranteeing driving safety.

Description

Technical Field

[0001] This invention relates to the field of battery self-discharge monitoring and early warning, and more particularly to a method, system, device and medium for monitoring and early warning of abnormal battery self-discharge. Background Technology

[0002] With the increasing severity of environmental pollution and energy shortages, traditional gasoline-powered vehicles, which suffer from heavy emissions and high energy consumption, have raised growing concerns. Electric vehicles, with their energy-saving and environmentally friendly advantages, have gradually gained global attention. Lithium-ion batteries, characterized by high power density, long lifespan, low maintenance costs, and zero pollution, are widely used as the power source for electric vehicles.

[0003] In existing electric vehicles, individual battery cells are often connected in series or parallel to provide power to the system. All individual battery cells within the system need to maintain good consistency throughout their entire lifespan. Poor consistency can lead to issues such as overcharging and over-discharging of individual cells, different battery degradation rates, and large temperature differences between cells, resulting in decreased system performance, shortened lifespan, and increased safety hazards. The main indicators for judging battery system consistency include capacity, internal resistance, self-discharge rate, and average discharge voltage. Among these, measuring the self-discharge rate is the most time-consuming, and self-discharge affects battery capacity and internal resistance. Therefore, achieving accurate and rapid measurement of the self-discharge rate is crucial for screening for system battery consistency. Furthermore, severe self-discharge can lead to thermal runaway of the battery, and the self-discharge rate is also an important reference indicator in battery safety management. However, current self-discharge detection methods have poor accuracy and are insufficient to meet practical application requirements. Summary of the Invention

[0004] In view of the problems existing in the prior art, the present invention proposes a method, system, device and medium for monitoring and early warning of abnormal battery self-discharge, which mainly solves the problem of inaccurate self-discharge detection in the existing technology.

[0005] To achieve the above and other objectives, the technical solution adopted by the present invention is as follows.

[0006] This application provides a method for monitoring and warning of abnormal battery self-discharge, comprising: collecting operational status data of a vehicle battery; inputting the operational status data into a battery self-discharge warning model to obtain battery status prediction information, wherein the battery self-discharge warning model is obtained by pre-training the model based on static and dynamic data, the static data being the correspondence between the sudden drop in battery terminal voltage and the self-discharge equivalent resistance value when the battery is in a static state, and the dynamic data being the difference in battery parameters between a normal battery and a self-discharge faulty battery under battery usage conditions; and matching a corresponding abnormality handling strategy according to the battery status prediction information.

[0007] In one embodiment of this application, the method for acquiring the static data includes: constructing a battery equivalent circuit model; applying a current pulse to the battery under test based on the battery equivalent circuit model; allowing the battery under test to rest after the current pulse and obtaining the voltage drop value of the battery under test at the last stage of resting, so as to calculate the corresponding self-discharge equivalent resistance and establish the corresponding relationship.

[0008] In one embodiment of this application, the step of setting the battery under test after a current pulse to rest and obtaining the voltage drop value of the battery under test during the final resting stage to calculate the corresponding self-discharge equivalent resistance and establish the correspondence includes: measuring the voltage change of the battery under test during the final resting stage; determining the self-discharge time constant of the battery under test based on the voltage change and compensating for the voltage drop caused by self-discharge during the overvoltage recovery period to obtain the voltage drop value; calculating the equivalent capacitance of the battery equivalent circuit model based on the voltage drop value; and determining the self-discharge equivalent resistance based on the equivalent capacitance and the time constant to establish the correspondence between the voltage drop value and the self-discharge equivalent resistance.

[0009] In one embodiment of this application, the dynamic data acquisition method includes: acquiring the terminal voltage of each battery cell in the battery under the use state; determining the suspected faulty battery cell based on the terminal voltage; comparing the battery parameters of the suspected faulty battery cell with the battery parameters of the other normal battery cells in the battery to complete the fault diagnosis and obtain the dynamic data.

[0010] In one embodiment of this application, the step of comparing the battery parameters of the suspected faulty battery cell with the battery parameters of the other normal battery cells in the battery includes: evaluating the battery parameters of the suspected faulty battery cell and the normal battery cells respectively using a joint algorithm of least squares method and extended Kalman filter method, so as to compare the corresponding battery parameters based on the evaluation results.

[0011] In one embodiment of this application, before performing battery parameter evaluation, the method further includes: calculating the average battery parameter value of each normal battery cell, so as to perform battery parameter evaluation based on the average battery parameter value.

[0012] In one embodiment of this application, the pre-training process of the battery self-discharge warning model includes: creating an initial model using a deep learning architecture; constructing a sample dataset based on the static data and the dynamic data, dividing the sample dataset into a training set, a validation set, and a test set according to time order; training the initial model and adjusting its parameters based on the training set and the validation set, and evaluating the performance of the trained model based on the test set, so as to use the model whose performance meets the preset conditions as the battery self-discharge warning model.

[0013] This application also provides a battery self-discharge abnormality monitoring and early warning system, comprising: a data acquisition module for acquiring vehicle battery operating status data; a status prediction module for inputting the operating status data into a battery self-discharge early warning model to obtain battery status prediction information, wherein the battery self-discharge early warning model is obtained by model pre-training based on static data and dynamic data, the static data being the correspondence between the battery terminal voltage drop value and the self-discharge equivalent resistance value in the battery's static state, and the dynamic data being the difference between battery parameters of a normal battery and a self-discharge faulty battery in the battery's usage state; and an anomaly handling module for matching a corresponding anomaly handling strategy according to the battery status prediction information.

[0014] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the battery self-discharge abnormal monitoring and early warning method.

[0015] This application also provides a readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the battery self-discharge abnormal monitoring and early warning method.

[0016] As described above, the battery self-discharge abnormal monitoring and early warning method, system, device and medium proposed in this application have the following beneficial effects.

[0017] This application constructs an electromagnetic self-discharge early warning model based on the correspondence between the battery terminal voltage and the self-discharge equivalent resistance value under static conditions, as well as the differences in battery parameters between normal batteries and self-discharge faulty batteries. By using this model to predict the battery status in real time, the accuracy of self-discharge anomaly monitoring and early warning can be effectively improved, ensuring driving safety. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a battery self-discharge abnormality monitoring and early warning method in one embodiment of this application.

[0019] Figure 2 This is a block diagram of a battery self-discharge abnormality monitoring and early warning system in one embodiment of this application.

[0020] Figure 3 This is a schematic diagram of the device in one embodiment of this application. Detailed Implementation

[0021] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0022] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0023] The inventor discovered through research that:

[0024] Current technologies for testing self-discharge anomalies in power batteries primarily rely on laboratory settings to allow the batteries to rest and measure voltage drop rates or capacity recovery rates. These methods are extremely time-consuming, with experimental periods ranging from one day to several weeks. Reducing measurement time in current-measuring scenarios requires expensive equipment. To shorten measurement time and save space and manpower, researchers have made numerous attempts. One approach is to accelerate the self-discharge rate by altering ambient temperature and battery state of charge (SOC), allowing for relatively large changes in measurement parameters within a shorter timeframe. While this method saves experimental time, it also accelerates battery aging and increases damage, making it suitable only for laboratory research and unsuitable for large-scale application in actual production.

[0025] Existing technology also discloses a method and device for detecting self-discharge of power batteries, a vehicle, and a storage medium. The method includes: acquiring a high-voltage power-on signal; recording the acquisition time T1 of the high-voltage power-on signal and recording the voltage value of each individual battery cell, calculating the average voltage Uavgd of each individual battery cell; acquiring a high-voltage power-on signal, recording the acquisition time T2 of the high-voltage power-on signal, acquiring the lowest individual battery cell voltage Umin, and acquiring the cell number n; acquiring the voltage value Un of the cell with cell number n; calculating the average voltage Uavgu; calculating the self-discharge parameter K; determining whether the self-discharge parameter K meets the alarm conditions, and if so, issuing an alarm; the self-discharge parameter can be calculated through a normal power-off and power-on process to characterize the normal self-discharge situation. However, due to factors such as the battery schedule and voltage fluctuations during vehicle operation, the self-discharge obtained by this method is not accurate in the frame preceding the high-voltage power-on signal.

[0026] Based on the problems existing in the prior art, this application proposes a method, system, device and medium for monitoring and early warning of abnormal battery self-discharge. The technical solution of this application will be described in detail below with reference to specific embodiments.

[0027] Please see Figure 1 , Figure 1 This is a flowchart illustrating a battery self-discharge anomaly monitoring and early warning method according to an embodiment of this application. The method provided in this embodiment includes the following steps:

[0028] Step S100: Collect the operating status data of the vehicle battery;

[0029] After the vehicle is powered on, the vehicle's battery management system can collect battery status data, including battery voltage, state of charge, and other battery-related parameters. Alternatively, other vehicle sensors can also collect battery status data; the specific devices used for data collection can be configured and adjusted according to actual application requirements. Data collection can continue throughout the entire vehicle's operation, or the collection interval can be set for periodic data collection to reduce system resource consumption. The data collection interval can also be adjusted based on the trend of the previous data collection. For example, if the trend of the previous data collection is close to an abnormal threshold, the sampling interval can be shortened. Of course, the specific sampling method can be adjusted according to actual application requirements, and there are no restrictions here.

[0030] Step S110: Input the operating status data into the battery self-discharge early warning model to obtain battery status prediction information. The battery self-discharge early warning model is obtained by pre-training the model based on static data and dynamic data. The static data is the correspondence between the battery terminal voltage drop value and the self-discharge equivalent resistance value when the battery is in a static state. The dynamic data is the difference between the battery parameters of a normal battery and a self-discharge faulty battery under battery usage conditions.

[0031] In one embodiment, a pre-trained battery self-discharge early warning model can be used to predict the battery status of the vehicle. The pre-training data of this model includes two parts: static data and dynamic data. The static data is acquired by: constructing a battery equivalent circuit model; applying a current pulse to the battery under test based on the battery equivalent circuit model; allowing the battery under test to rest after the current pulse and obtaining the voltage drop value of the battery under test at the last stage of rest to calculate the corresponding self-discharge equivalent resistance and establish the correspondence. Specifically, the battery equivalent circuit model can be an RC equivalent circuit, consisting of one or more resistors and capacitors forming an RC circuit connected to the battery under test as a power source. The specific battery equivalent circuit model can also be adjusted according to actual application requirements, and is not limited here. A short current pulse is applied to the battery under test connected to the battery equivalent circuit model, then the battery is allowed to rest, and the change in the voltage of the battery under test during the resting process is measured. The static data in this embodiment only considers the dominant reaction at each stage of the resting process, decoupling the complex reaction mechanism and reducing the computational load while shortening the measurement time. Specifically, the device plays a dominant role in the recovery of overvoltage during the initial resting period, while the self-discharge of the battery becomes dominant during the final resting period. Therefore, the self-discharge equivalent resistance of the battery under test can be determined by the data from the final resting period (i.e., the last stage of resting).

[0032] In one embodiment, the step of allowing the battery under test (BUT) to rest after a current pulse and obtaining the voltage drop at the end of the resting period to calculate the corresponding self-discharge equivalent resistance and establish the correspondence includes: measuring the voltage change of the BUT during the final resting period; determining the self-discharge time constant of the BUT based on the voltage change and compensating for the voltage drop caused by self-discharge during the overvoltage recovery period to obtain the voltage drop; calculating the equivalent capacitance of the battery's equivalent circuit model based on the voltage drop; and determining the self-discharge equivalent resistance based on the equivalent capacitance and the time constant to establish the correspondence between the voltage drop and the self-discharge equivalent resistance. Specifically, the voltage change during the final resting period can be measured, and the voltage drop over the entire time range of the final period can be calculated based on the voltage change. Then, the final voltage drop is obtained by adjusting the step size for the self-discharge voltage drop during the voltage recovery period in the early stage of resting. Furthermore, the self-discharge time constant can be calculated based on the voltage change. The self-discharge time constant, also known as the time constant, is an important parameter describing the capacitor discharge process. It reflects the rate at which a capacitor discharges through a resistor, and is the product of resistance (R) and capacitance (C), i.e., T = RC. This formula reveals the direct relationship between the capacitor's discharge time and its resistance and capacitance. Specifically, the time constant T represents a portion of the time it takes for the capacitor voltage to drop from its initial value to its final stable value, typically referring to the time required for the capacitor voltage to drop to 37% of its initial value. After calculating the equivalent capacitance of the battery under test in the battery's equivalent circuit model based on the voltage drop value, the self-discharge equivalent resistance of the battery under test can be obtained by solving for the correspondence between the equivalent capacitance and the self-discharge time constant. This establishes the correspondence between the voltage drop value of the battery under test and the self-discharge equivalent resistance, serving as static data. Static data from multiple batteries under test are then combined to form a static sample dataset.

[0033] In one embodiment, the dynamic data acquisition method includes: acquiring the terminal voltage of each battery cell in the battery under usage conditions; identifying suspected faulty battery cells based on the terminal voltage; and comparing the battery parameters of the suspected faulty battery cells with the battery parameters of the remaining normal battery cells in the battery to complete fault diagnosis and obtain the dynamic data. Specifically, the point feature of a large difference between the battery's self-discharge triggering terminal voltage and that of a normal battery can be utilized, and a time series correlation algorithm can be used to perform online diagnosis of the faulty battery. A diagnostic threshold range for suspected self-discharge faults can be preset, which can be the range of terminal voltages of battery cells suspected of experiencing self-discharge faults. By collecting the terminal voltages of each battery cell in the vehicle battery, battery cells with terminal voltages exceeding the preset diagnostic threshold range are marked as suspected faulty battery cells, while the remaining cells are normal battery cells. All normal battery cells in the vehicle battery can be converted into an "average battery," and the battery parameters of the suspected faulty battery cells and the normal battery cells can be evaluated using a joint algorithm of least squares method and extended Kalman filter method, respectively, so as to perform corresponding battery parameter comparison based on the evaluation results. The joint algorithm of least squares and extended Kalman filtering is a method that combines the estimation techniques of least squares and extended Kalman filtering to address state estimation problems in nonlinear systems. This joint algorithm is typically applied to complex systems requiring high-precision state estimation. Least squares is a commonly used estimation method that estimates unknown parameters by minimizing the sum of squared errors. This method is suitable for linear systems by constructing a matrix X containing all possible eigenvalues ​​and a target variable vector y, using the formula b = (X... T X) -1 X T The parameters are estimated using the least squares method. However, for nonlinear systems, the least squares method may not accurately estimate the node positions, which necessitates the use of the extended Kalman filter (EKF). The EKF is an extension of the Kalman filter specifically designed for state estimation problems in nonlinear systems. EKF expands the application range of the Kalman filter by linearizing the nonlinear system. It consists of two main steps: a prediction step and an update step. The prediction step predicts the state and estimation error covariance at the next time step based on the system model, while the update step updates the state estimate and estimation error covariance based on the observations and the system model. EKF approximates the state transition and observation equations by calculating the Jacobian matrix of the nonlinear function, which is crucial for handling nonlinear problems. In the joint algorithm, the least squares method linearly fits the battery parameters, while the extended Kalman filter estimates the parameter nodes in the fitted curve.

[0034] In one embodiment, before performing battery parameter evaluation, the method further includes: calculating the average battery parameters of each normal battery cell, and then evaluating the battery parameters based on the average battery parameters. Alternatively, before estimating battery parameters using a joint algorithm, the average value of each normal battery cell can be calculated first, and this average value can be used as the battery parameters of the "average battery." Then, the joint algorithm is used to estimate the corresponding battery parameters at different times. The battery parameters of the suspected faulty battery cell are compared with the corresponding battery parameters of the normal battery. Differences in battery parameters can be compared, such as the difference in state of charge (SOC). When the difference between the two battery parameters exceeds a preset self-discharge fault threshold, the suspected faulty battery cell can be determined as a faulty battery cell, and the difference between the normal battery cell and the self-discharge faulty battery cell is recorded as dynamic data.

[0035] In one embodiment, the pre-training process of the battery self-discharge early warning model includes: creating an initial model using a deep learning architecture; constructing a sample dataset based on the static and dynamic data, and dividing the sample dataset into a training set, a validation set, and a test set according to time sequence; training and adjusting the model parameters based on the training set and the validation set, and evaluating the performance of the trained model based on the test set, so that the model whose performance meets preset conditions is used as the battery self-discharge early warning model. Specifically, the dataset can be split according to the time sequence of the samples in the sample dataset, and the ratio of the training set, validation set, and test set can be set and adjusted according to actual application needs. The deep learning architecture can adopt network architectures such as recurrent neural networks, and can also be selected according to actual application needs, without limitation. Dividing the dataset based on time sequence, and using newer data for the validation set and test set, can ensure that the obtained model is closer to the state of the battery, ensuring the prediction effect.

[0036] Step S120: Match the corresponding anomaly handling strategy based on the battery state prediction information.

[0037] In one embodiment, battery state prediction information obtained from the input model state prediction can be matched with different degrees of self-discharge faults. Anomaly handling strategies corresponding to different degrees of self-discharge faults can be pre-stored in a database, and these strategies may include continuous monitoring, vehicle recall, etc. When the battery state prediction information determines that the vehicle has a mild self-discharge fault, the corresponding battery state can be continuously monitored, and relevant information can be displayed or alerted via the vehicle-side display device. If it is a severe self-discharge fault, the fault information is sent to the cloud, and a vehicle recall is triggered by contacting the maintenance center. Alternatively, the information can be displayed as an alarm on the vehicle-side display device or sent to a specific terminal for early warning. Specific anomaly handling strategies can be selected and adjusted according to actual application needs, and are not limited here.

[0038] Based on the technical solutions provided in the embodiments of this application above, a full life cycle self-discharge dataset can be constructed using static and dynamic data. The trained model can effectively extract the complex features of the battery, achieve accurate prediction of self-discharge throughout the entire life cycle, and predict self-discharge anomalies before the user perceives the anomalies, outputting relevant prompt information to avoid thermal runaway problems caused by self-discharge anomalies, ensuring driving safety, and enhancing the user experience.

[0039] Please see Figure 2 , Figure 2 This is a schematic diagram of a vehicle gear redundancy control system according to an embodiment of this application. The vehicle gear redundancy control system includes: a data acquisition module 20 for acquiring operating status data of the vehicle battery; a status prediction module 21 for inputting the operating status data into a battery self-discharge early warning model to obtain battery status prediction information, wherein the battery self-discharge early warning model is obtained through model pre-training based on static and dynamic data, the static data being the correspondence between the battery terminal voltage drop value and the self-discharge equivalent resistance value in the battery's static state, and the dynamic data being the difference in battery parameters between a normal battery and a self-discharge faulty battery in the battery's usage state; and an anomaly handling module 22 for matching a corresponding anomaly handling strategy based on the battery status prediction information.

[0040] In one embodiment, the state prediction module 21 is further used to build a battery equivalent circuit model; apply a current pulse to the battery under test based on the battery equivalent circuit model; let the battery under test stand after the current pulse and obtain the terminal voltage drop value of the battery under test in the last stage of standing, so as to calculate the corresponding self-discharge equivalent resistance and establish the correspondence.

[0041] In one embodiment, the state prediction module 21 is further configured to allow the battery under test to rest after the current pulse and obtain the voltage drop value of the battery under test in the final stage of resting, so as to calculate the corresponding self-discharge equivalent resistance. The step of establishing the correspondence includes: measuring the voltage change of the battery under test in the final stage of resting; determining the self-discharge time constant of the battery under test according to the voltage change, and compensating for the voltage drop caused by self-discharge during the overvoltage recovery period to obtain the voltage drop value; calculating the equivalent capacitance of the battery equivalent circuit model according to the voltage drop value; and determining the self-discharge equivalent resistance based on the equivalent capacitance and the time constant to establish the correspondence between the voltage drop value and the self-discharge equivalent resistance.

[0042] In one embodiment, the state prediction module 21 is further configured to acquire the dynamic data by: acquiring the terminal voltage of each battery cell in the battery under the usage state; determining the suspected faulty battery cell based on the terminal voltage; comparing the battery parameters of the suspected faulty battery cell with the battery parameters of the remaining normal battery cells in the battery to complete the fault diagnosis and obtain the dynamic data.

[0043] In one embodiment, the state prediction module 21 is further configured to compare the battery parameters of the suspected faulty battery cell with the battery parameters of the other normal battery cells in the battery, including: evaluating the battery parameters of the suspected faulty battery cell and the normal battery cells respectively using a joint algorithm of least squares method and extended Kalman filter method, so as to compare the corresponding battery parameters based on the evaluation results.

[0044] In one embodiment, the state prediction module 21 is further configured to: calculate the average battery parameter of each of the normal battery cells before performing battery parameter evaluation, so as to perform battery parameter evaluation based on the average battery parameter.

[0045] In one embodiment, the state prediction module 21 is further configured to create an initial model using a deep learning architecture; construct a sample dataset based on the static data and the dynamic data, and divide the sample dataset into a training set, a validation set, and a test set according to time order; train the initial model and adjust the model parameters based on the training set and the validation set, and evaluate the performance of the trained model based on the test set, so as to use the model whose performance meets the preset conditions as the battery self-discharge warning model.

[0046] The aforementioned battery self-discharge abnormal monitoring and early warning method can be implemented in the form of a computer program, which can be implemented in, for example... Figure 3 The computer device shown runs on the computer. The computer device includes: memory, processor, and computer programs stored in the memory and executable on the processor.

[0047] Each module in the aforementioned battery self-discharge anomaly monitoring and early warning method can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the terminal's memory in hardware form, or stored in the terminal's memory in software form, so that the processor can call and execute the corresponding operations of each module. The processor can be a central processing unit (CPU), microprocessor, microcontroller, etc.

[0048] like Figure 3The diagram shown is a schematic representation of the internal structure of a computer device in one embodiment. A computer device is provided, comprising: collecting operational status data of a vehicle battery; inputting the operational status data into a battery self-discharge early warning model to obtain battery status prediction information, wherein the battery self-discharge early warning model is obtained through model pre-training based on static and dynamic data, the static data being the correspondence between the battery's terminal voltage drop value and the self-discharge equivalent resistance value in a static state, and the dynamic data being the difference in battery parameters between a normal battery and a self-discharge faulty battery in a used state; and matching a corresponding anomaly handling strategy based on the battery status prediction information.

[0049] In one embodiment, when the processor executes the above-mentioned method, the method for acquiring the static data includes: building a battery equivalent circuit model; applying a current pulse to the battery under test based on the battery equivalent circuit model; allowing the battery under test to rest after the current pulse and acquiring the terminal voltage drop value of the battery under test in the final stage of resting, so as to calculate the corresponding self-discharge equivalent resistance and establish the corresponding relationship.

[0050] In one embodiment, when the processor executes the above-mentioned steps, the steps of resting the battery under test after the current pulse and obtaining the voltage drop value of the battery under test in the final resting stage to calculate the corresponding self-discharge equivalent resistance and establish the correspondence include: measuring the voltage change of the battery under test in the final resting stage; determining the self-discharge time constant of the battery under test based on the voltage change, and compensating for the voltage drop caused by self-discharge during the overvoltage recovery period to obtain the voltage drop value; calculating the equivalent capacitance of the battery equivalent circuit model based on the voltage drop value; and determining the self-discharge equivalent resistance based on the equivalent capacitance and the time constant to establish the correspondence between the voltage drop value and the self-discharge equivalent resistance.

[0051] In one embodiment, when the processor executes the above-mentioned method, the acquisition of the dynamic data includes: acquiring the terminal voltage of each battery cell in the battery under the use state; determining the suspected faulty battery cell based on the terminal voltage; comparing the battery parameters of the suspected faulty battery cell with the battery parameters of the other normal battery cells in the battery to complete the fault diagnosis and obtain the dynamic data.

[0052] In one embodiment, when the processor executes the above-mentioned step of comparing the battery parameters of the suspected faulty battery cell with the battery parameters of the other normal battery cells in the battery, the steps include: evaluating the battery parameters of the suspected faulty battery cell and the normal battery cells respectively using a joint algorithm of least squares method and extended Kalman filter method, so as to compare the corresponding battery parameters based on the evaluation results.

[0053] In one embodiment, when the processor executes the above-mentioned process, the process before performing battery parameter evaluation further includes: calculating the average battery parameter value of each of the normal battery cells, so as to perform battery parameter evaluation based on the average battery parameter value.

[0054] In one embodiment, when the processor executes the above-mentioned process, the pre-training process of the battery self-discharge warning model includes: creating an initial model through a deep learning architecture; constructing a sample dataset based on the static data and the dynamic data, dividing the sample dataset into a training set, a validation set, and a test set according to time order; training the initial model and adjusting the model parameters based on the training set and the validation set, and evaluating the performance of the trained model based on the test set, so as to use the model whose performance meets the preset conditions as the battery self-discharge warning model.

[0055] In one embodiment, the aforementioned computer device can be used as a server, including but not limited to a standalone physical server or a server cluster consisting of multiple physical servers. The computer device can also be used as a terminal, including but not limited to mobile phones, tablets, personal digital assistants, or smart devices. Figure 3 As shown, the computer device includes a processor, non-volatile storage medium, internal memory, display screen, and network interface connected via a system bus.

[0056] The processor of this computer device provides computing and control capabilities to support the operation of the entire device. The non-volatile storage medium of the computer device stores the operating system and computer programs. These programs can be executed by the processor to implement the battery self-discharge abnormality monitoring and early warning method provided in the above embodiments. The internal memory of the computer device provides a cached operating environment for the operating system and computer programs stored in the non-volatile storage medium. The display interface can display data via a screen. The screen can be a touchscreen, such as a capacitive or electronic screen, and can generate corresponding instructions by receiving click operations on the controls displayed on the touchscreen.

[0057] Those skilled in the art will understand that Figure 3 The structure of the computer device shown in the figure is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0058] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it performs the following steps: collecting operational status data of a vehicle battery; inputting the operational status data into a battery self-discharge early warning model to obtain battery status prediction information, wherein the battery self-discharge early warning model is obtained by model pre-training based on static and dynamic data, the static data being the correspondence between the battery terminal voltage drop value and the self-discharge equivalent resistance value in the battery's static state, and the dynamic data being the difference in battery parameters between a normal battery and a self-discharge faulty battery in the battery's usage state; and matching a corresponding anomaly handling strategy according to the battery status prediction information.

[0059] In one embodiment, when the computer program is executed by the processor, the method for acquiring the static data includes: building a battery equivalent circuit model; applying a current pulse to the battery under test based on the battery equivalent circuit model; allowing the battery under test to rest after the current pulse and acquiring the voltage drop value of the battery under test at the last stage of resting, so as to calculate the corresponding self-discharge equivalent resistance and establish the correspondence.

[0060] In one embodiment, when the computer program is executed by the processor, the steps of resting the battery under test after a current pulse and obtaining the voltage drop value of the battery under test during the final resting stage to calculate the corresponding self-discharge equivalent resistance and establish the correspondence include: measuring the voltage change of the battery under test during the final resting stage; determining the self-discharge time constant of the battery under test based on the voltage change and compensating for the voltage drop caused by self-discharge during the overvoltage recovery period to obtain the voltage drop value; calculating the equivalent capacitance of the battery equivalent circuit model based on the voltage drop value; and determining the self-discharge equivalent resistance based on the equivalent capacitance and the time constant to establish the correspondence between the voltage drop value and the self-discharge equivalent resistance.

[0061] In one embodiment, when the computer program is executed by the processor, the method for acquiring the dynamic data includes: acquiring the terminal voltage of each battery cell in the battery under the use state; determining the suspected faulty battery cell based on the terminal voltage; comparing the battery parameters of the suspected faulty battery cell with the battery parameters of the other normal battery cells in the battery to complete the fault diagnosis and obtain the dynamic data.

[0062] In one embodiment, when the computer program is executed by the processor, the step of comparing the battery parameters of the suspected faulty battery cell with the battery parameters of the other normal battery cells in the battery includes: evaluating the battery parameters of the suspected faulty battery cell and the normal battery cells respectively using a joint algorithm of least squares method and extended Kalman filter method, so as to compare the corresponding battery parameters based on the evaluation results.

[0063] In one embodiment, when the computer program is executed by a processor, the implementation of the battery parameter evaluation further includes: calculating the average battery parameter of each of the normal battery cells, so as to perform the battery parameter evaluation based on the average battery parameter.

[0064] In one embodiment, when the computer program is executed by the processor, the pre-training process of the battery self-discharge warning model includes: creating an initial model using a deep learning architecture; constructing a sample dataset based on the static data and the dynamic data, dividing the sample dataset into a training set, a validation set, and a test set according to time order; training the initial model and adjusting its parameters based on the training set and the validation set, and evaluating the performance of the trained model based on the test set, so as to use the model whose performance meets the preset conditions as the battery self-discharge warning model.

[0065] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), etc.

[0066] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for monitoring and early warning of abnormal battery self-discharge, characterized in that, include: Collect vehicle battery operating status data; The operating status data is input into the battery self-discharge early warning model to obtain battery status prediction information. The battery self-discharge early warning model is obtained through model pre-training based on static and dynamic data. The static data is the correspondence between the battery's terminal voltage drop value and the self-discharge equivalent resistance value in a static state. The dynamic data is the difference in battery parameters between a normal battery and a battery with a self-discharge fault in a used state. The static data is obtained by: constructing a battery equivalent circuit model; applying a current pulse to the battery under test based on the battery equivalent circuit model; allowing the battery under test to rest after the current pulse and obtaining the terminal voltage drop value of the battery under test in the final stage of rest to calculate the corresponding self-discharge equivalent resistance and establish the correspondence. The corresponding anomaly handling strategy is matched based on the battery state prediction information.

2. The battery self-discharge abnormal monitoring and early warning method according to claim 1, characterized in that, The steps of allowing the battery under test to rest after being subjected to a current pulse and obtaining the voltage drop at the end of the resting period to calculate the corresponding self-discharge equivalent resistance and establish the corresponding relationship include: Measure the voltage change of the battery under test during the final stage of resting; The self-discharge time constant of the battery under test is determined based on the voltage change, and the voltage drop caused by self-discharge during the overvoltage recovery period is compensated to obtain the terminal voltage drop value. Calculate the equivalent capacitance of the battery equivalent circuit model based on the voltage drop value at the terminal. The self-discharge equivalent resistance is determined based on the equivalent capacitance and the time constant, so as to establish the correspondence between the terminal voltage drop value and the self-discharge equivalent resistance.

3. The battery self-discharge abnormal monitoring and early warning method according to claim 1, characterized in that, The methods for acquiring the dynamic data include: Obtain the terminal voltage of each individual cell in the battery while it is in use; The suspected faulty battery cell is determined based on the terminal voltage; The battery parameters of the suspected faulty battery cell are compared with the battery parameters of the other normal battery cells in the battery to complete the fault diagnosis and obtain the dynamic data.

4. The battery self-discharge abnormal monitoring and early warning method according to claim 3, characterized in that, The step of comparing the battery parameters of the suspected faulty battery cell with the battery parameters of the other normal battery cells in the battery includes: The battery parameters of the suspected faulty battery cells and the normal battery cells are evaluated using a joint algorithm of least squares method and extended Kalman filter method, respectively, and the corresponding battery parameters are compared based on the evaluation results.

5. The battery self-discharge abnormal monitoring and early warning method according to claim 4, characterized in that, Before performing battery parameter evaluation, the method further includes: calculating the average battery parameter value of each of the normal battery cells, so as to perform battery parameter evaluation based on the average battery parameter value.

6. The battery self-discharge abnormal monitoring and early warning method according to claim 1, characterized in that, The pre-training process of the battery self-discharge early warning model includes: Create an initial model using a deep learning architecture; A sample dataset is constructed based on the static data and the dynamic data, and the sample dataset is divided into a training set, a validation set and a test set according to the time sequence. The initial model is trained and its parameters are adjusted based on the training set and the validation set, and the performance of the trained model is evaluated based on the test set, so that the model whose performance meets the preset conditions is used as the battery self-discharge early warning model.

7. A battery self-discharge abnormality monitoring and early warning system, characterized in that, include: The data acquisition module is used to collect the operating status data of the vehicle battery; The state prediction module is used to input the operating state data into the battery self-discharge early warning model to obtain battery state prediction information. The battery self-discharge early warning model is obtained through model pre-training based on static and dynamic data. The static data is the correspondence between the battery's terminal voltage drop value and the self-discharge equivalent resistance value in a static state. The dynamic data is the difference in battery parameters between a normal battery and a self-discharge faulty battery in use. The static data is acquired by: building a battery equivalent circuit model; applying a current pulse to the battery under test based on the battery equivalent circuit model; allowing the battery under test to rest after the current pulse and obtaining the terminal voltage drop value of the battery under test in the final stage of resting, to calculate the corresponding self-discharge equivalent resistance and establish the correspondence. An anomaly handling module is used to match a corresponding anomaly handling strategy based on the battery state prediction information.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the battery self-discharge abnormal monitoring and early warning method as described in any one of claims 1 to 6.

9. A readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the battery self-discharge abnormal monitoring and early warning method as described in any one of claims 1 to 6.

Images