A method, device, equipment and storage medium for determining battery health status

The battery data is analyzed through the prediction model trained by Bayesian network, which solves the problem of insufficient accuracy in estimation of battery health status in the prior art, and achieves higher accuracy in determining health status.

CN117761563BActive Publication Date: 2024-11-08GUANGDONG POWER GRID CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410143178.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-01-31
Publication Date
2024-11-08
Estimated Expiration
2044-01-31

AI Technical Summary

Technical Problem

The prior art lacks accuracy and adaptability to estimating battery health status, resulting in low accuracy in determining battery health status.

Method used

Using Bayesian network training prediction model, analyses are constructed and model parameters are updated by acquiring multiple parameters in the battery data (such as voltage, current, temperature and cycle times) to construct a Bayesian network and update the model parameters, combining a priori distribution and likelihood The function performs posterior distribution calculation to achieve accurate prediction of the battery health status.

Benefits of technology

It improves the accuracy of battery health status determination, enables more comprehensive analysis of battery data, and provides higher accuracy in health status estimation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117761563B_ABST
    Figure CN117761563B_ABST
Patent Text Reader

Abstract

The present invention discloses a method, device, equipment and storage medium for determining the health status of a battery, and relates to the field of computer technology. The method comprises: obtaining battery data of a battery to be tested, the battery data comprising a plurality of battery parameters; inputting the battery data into a pre-trained prediction model, so that the prediction model analyzes the health status of the battery to be tested according to the battery data, and determines the health status of the battery to be tested. In the above technical scheme, after obtaining the battery data of the battery to be tested, the battery data of the battery to be tested can be input into a pre-trained prediction model, and the prediction model can analyze the health status of the battery to be tested according to the battery data of the battery to be tested. Since the prediction model is obtained by Bayesian network training, when the prediction model analyzes the health status of the battery to be tested according to the battery data of the battery to be tested, the battery data of the battery to be tested can be analyzed more comprehensively, so as to achieve a higher accuracy in determining the health status of the battery to be tested.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of computer technology, and in particular, to a method, apparatus, device, and storage medium for determining a health status of a battery. Background Art

[0002] Batteries are widely used as power storage devices in various fields, for example, in mobile devices, electric vehicles, renewable energy systems, etc. Battery state of health (SOH) estimation is crucial to ensure the performance, safety, and life of the battery.

[0003] In recent years, the development of machine learning and deep learning technologies has provided new opportunities for battery health state estimation. In existing technologies, battery SOH estimation is usually based on parameters such as battery cycle number, capacity decay rate, and internal resistance, combined with deep learning technology.

[0004] However, the existing technology for battery SOH estimation is often limited by the accuracy and adaptability of the model, resulting in low accuracy in determining the health status of the battery. Summary of the invention

[0005] The present invention provides a method, device, equipment and storage medium for determining the health state of a battery, so as to accurately determine the health state of the battery.

[0006] In a first aspect, an embodiment of the present invention provides a method for determining a health status of a battery, comprising:

[0007] Acquire battery data of a battery to be tested, wherein the battery data includes a plurality of battery parameters;

[0008] The battery data is input into a pre-trained prediction model, so that the prediction model analyzes the health state of the battery to be tested according to the battery data to determine the health state of the battery to be tested.

[0009] The technical solution of an embodiment of the present invention provides a method for determining the health status of a battery, comprising: obtaining battery data of a battery to be tested, wherein the battery data includes multiple battery parameters; inputting the battery data into a pre-trained prediction model, so that the prediction model analyzes the health status of the battery to be tested according to the battery data, and determines the health status of the battery to be tested. In the above technical solution, after obtaining the battery data of the battery to be tested, the battery data of the battery to be tested can be input into a pre-trained prediction model, and the prediction model can analyze the health status of the battery to be tested according to the battery data of the battery to be tested. Since the prediction model is obtained by Bayesian network training, when the prediction model analyzes the health status of the battery to be tested according to the battery data of the battery to be tested, it can more comprehensively analyze the battery data of the battery to be tested, thereby determining the health status of the battery to be tested with higher accuracy.

[0010] Furthermore, the battery parameters include at least voltage, current, temperature and cycle number.

[0011] Furthermore, the training process of the prediction model includes:

[0012] Acquire training battery data and the real health status corresponding to each training battery data, wherein the training battery data includes a plurality of training battery parameters;

[0013] The training battery parameters are used as nodes, and the predicted health status corresponding to the training battery data and the model parameters of the prediction model are used as target nodes to construct a Bayesian network;

[0014] The training battery data and the real health status corresponding to the training battery data are input into the neural network model as training data, the neural network model is trained, and the model parameters of the neural network model are updated based on the Bayesian network to obtain the prediction model.

[0015] Furthermore, after constructing the Bayesian network by taking each of the training battery parameters as a node, the predicted health state corresponding to the training battery data and the model parameters of the prediction model as the target node, the method further includes:

[0016] determining a prior distribution of each of the training battery parameters to define a first likelihood function for the predicted health state;

[0017] For each of the training battery parameters, combining the prior distribution of each of the training battery parameters, the first likelihood function and the current training battery parameters, determining a first posterior distribution corresponding to the predicted health state;

[0018] The Bayesian network is updated according to the first posterior distribution.

[0019] Furthermore, after constructing the Bayesian network by taking each of the training battery parameters as a node, the predicted health state corresponding to the training battery data and the model parameters of the prediction model as the target node, the method further includes:

[0020] defining a second likelihood function for model parameters of the prediction model;

[0021] For each of the training battery parameters, determine a second posterior distribution corresponding to each of the model parameters in combination with the prior distribution of each of the training battery parameters, the second likelihood function, the current training battery parameters and the current health status;

[0022] The Bayesian network is updated according to the second posterior distribution.

[0023] Furthermore, each of the training battery data and the real health status corresponding to each of the training battery data are input into a neural network model as training data, the neural network model is trained, and the model parameters of the neural network model are updated based on the Bayesian network to obtain the prediction model, including:

[0024] Inputting each of the training battery data and the real health status corresponding to each of the training battery data as training data into a neural network model to train the neural network model;

[0025] The model parameters of the neural network model are updated according to the training data and the second posterior distribution corresponding to each of the model parameters to obtain the prediction model.

[0026] Furthermore, before inputting each of the training battery data and the real health status corresponding to each of the training battery data into the neural network model as training data, the method further includes:

[0027] When it is determined that any of the training battery parameters does not conform to the prior distribution of the training battery parameters, the training battery parameters are adjusted based on the prior distribution of the training battery parameters.

[0028] In a second aspect, an embodiment of the present invention further provides a device for determining a health status of a battery, comprising:

[0029] An acquisition module, used to acquire battery data of a battery to be tested, wherein the battery data includes a plurality of battery parameters;

[0030] The analysis module is used to input the battery data into a pre-trained prediction model so that the prediction model analyzes the health state of the battery to be tested according to the battery data to determine the health state of the battery to be tested.

[0031] In a third aspect, an embodiment of the present invention further provides a computer device, the computer device comprising:

[0032] at least one processor; and a memory communicatively coupled to the at least one processor;

[0033] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the battery health status determination method as described in any one of the first aspects.

[0034] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, are used to execute a method for determining a battery health status as described in any one of the first aspects.

[0035] In a fifth aspect, the present application provides a computer program product, which includes computer instructions. When the computer instructions are executed on a computer, the computer executes the method for determining the health status of a battery as provided in the first aspect.

[0036] It should be noted that the above-mentioned computer instructions may be stored in whole or in part on a computer-readable storage medium. The computer-readable storage medium may be packaged together with the processor of the battery health status determination device, or may be packaged separately from the processor of the battery health status determination device, and this application does not limit this.

[0037] The description of the second, third, fourth and fifth aspects in this application can refer to the detailed description of the first aspect; and the beneficial effects of the description of the second, third, fourth and fifth aspects can refer to the beneficial effect analysis of the first aspect, which will not be repeated here.

[0038] In this application, the name of the battery health status determination device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of the claims of this application and their equivalent technologies.

[0039] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1 A flowchart of a method for determining a battery health status provided by an embodiment of the present invention;

[0042] Figure 2 A flowchart of another method for determining the health status of a battery provided by an embodiment of the present invention;

[0043] Figure 3 A schematic diagram of a neural network model in another method for determining a battery health state provided by an embodiment of the present invention;

[0044] Figure 4 A schematic diagram of the structure of a device for determining the health status of a battery provided by an embodiment of the present invention;

[0045] Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0047] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0048] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0049] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0050] It should be mentioned before discussing the exemplary embodiments in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to methods, functions, procedures, subroutines, subprograms, etc. In addition, the embodiments in the present invention and the features in the embodiments can be combined with each other without conflict.

[0051] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0052] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0053] Figure 1 A flowchart of a method for determining the health status of a battery provided in an embodiment of the present invention is provided. This embodiment is applicable to situations where the health status of a battery needs to be accurately determined. The method can be executed by a battery health status determination device, such as Figure 1 As shown, the specific steps include:

[0054] Step 110: Obtain battery data of the battery to be tested.

[0055] The battery data includes a plurality of battery parameters.

[0056] The battery to be tested can be understood as a battery that needs to be estimated for SOH. The battery data of the battery to be tested can be understood as data indicating the operating status of the battery to be tested. The battery data includes multiple battery parameters, and the battery parameters may include voltage, current, temperature, number of cycles, etc.

[0057] Specifically, during the operation of the battery to be tested, battery parameters such as voltage, current, temperature, and number of cycles of the battery to be tested can be obtained to obtain battery data of the battery to be tested.

[0058] Step 120: input the battery data into a pre-trained prediction model, so that the prediction model analyzes the health status of the battery to be tested according to the battery data to determine the health status of the battery to be tested.

[0059] The prediction model is pre-trained based on training battery data including multiple training battery parameters and the actual health status corresponding to each training battery data, and can be used to determine the SOH of the battery to be tested based on the battery data of the battery to be tested.

[0060] In the process of training the neural network model based on the training battery data and the actual SOH corresponding to each training battery data, the model parameters can be updated based on the Bayesian network.

[0061] Specifically, the battery data of the battery to be tested is input into a pre-trained prediction model, and the prediction model can analyze the health status of the battery to be tested based on the battery data of the battery to be tested to obtain the SOH of the battery to be tested. The present application uses a prediction model obtained by Bayesian network training to analyze the SOH of the battery to be tested based on the battery data of the battery to be tested, which can process the battery data more comprehensively and achieve a more accurate determination of the SOH of the battery to be tested.

[0062] In an embodiment of the present invention, after obtaining the battery data of the battery to be tested, the battery data of the battery to be tested can be input into a pre-trained prediction model. The prediction model can analyze the health status of the battery to be tested based on the battery data of the battery to be tested, thereby determining a more accurate health status of the battery to be tested.

[0063] The method for determining the health status of a battery provided in an embodiment of the present invention comprises: obtaining battery data of a battery to be tested, wherein the battery data comprises a plurality of battery parameters; inputting the battery data into a pre-trained prediction model, so that the prediction model analyzes the health status of the battery to be tested according to the battery data, and determines the health status of the battery to be tested. In the above technical scheme, after obtaining the battery data of the battery to be tested, the battery data of the battery to be tested can be input into a pre-trained prediction model, and the prediction model can analyze the health status of the battery to be tested according to the battery data of the battery to be tested. Since the prediction model is obtained by Bayesian network training, when the prediction model analyzes the health status of the battery to be tested according to the battery data of the battery to be tested, the battery data of the battery to be tested can be analyzed more comprehensively, so as to determine the health status of the battery to be tested with higher accuracy.

[0064] Figure 2 This is a flow chart of another method for determining the health status of a battery provided by an embodiment of the present invention. This embodiment is specific based on the above embodiment. Figure 2 As shown, in this embodiment, the method may further include:

[0065] Step 210: Obtain training battery data and the actual health status corresponding to each training battery data.

[0066] The training battery data includes a plurality of training battery parameters, and the training battery parameters include at least voltage, current, temperature and number of cycles.

[0067] The training battery data is battery data of a training battery. The training battery can be understood as a battery of the same model as the battery to be tested.

[0068] Specifically, based on at least one training battery, training battery data and the real health status corresponding to each training battery data are obtained. Specifically, multiple training battery parameters of each training battery can be obtained, for example, at least the pack voltage, current, temperature and number of cycles of each training battery are obtained.

[0069] In practical applications, after obtaining the training battery data and the actual health status corresponding to each training battery data, the data needs to be preprocessed. For example, the data can be cleaned, normalized, and sequenced to ensure that the data is suitable for the input requirements of the prediction model.

[0070] Step 220: construct a Bayesian network by taking each of the training battery parameters as a node, and the predicted health status corresponding to the training battery data and the model parameters of the prediction model as target nodes.

[0071] The Bayesian network may be a variable parameter Bayesian network (VP-BN), and the Bayesian network is composed of multiple nodes and causal connections between the nodes.

[0072] Specifically, the battery data of the battery determines the health state of the battery, and the parameters of the Bayesian network determine the parameters of the prediction model. The parameters of the Bayesian network are determined by the battery data of the battery and the health state of the battery. Therefore, the training battery parameters contained in the training battery data can be used as nodes, and the predicted health state corresponding to the training battery data and the model parameters of the prediction model can be used as target nodes to construct a Bayesian network.

[0073] Of course, when the model parameters of the prediction model are target nodes, the predicted health status corresponding to the training battery data can also be used as a node to update the model parameters of the prediction model.

[0074] In the embodiment of the present invention, the Bayesian network is constructed by taking each training battery parameter as a node and the predicted health state and model parameters of the prediction model corresponding to the training battery data as the target node.

[0075] Step 230, determine the prior distribution of each of the training battery parameters, and define a first likelihood function for the predicted health state; for each of the training battery parameters, determine a first posterior distribution corresponding to the predicted health state in combination with the prior distribution of each of the training battery parameters, the first likelihood function and the current training battery parameters; and update the Bayesian network according to the first posterior distribution.

[0076] Specifically, we can first use the formula Determine the prior distribution of each training battery parameter, and then define the first likelihood function for the health state To represent the probability that the health state is SOH under given battery parameters. i represents the training battery parameters, μ i represents the mean of the training battery parameters, represents the variance of the training battery parameters, μ SOH represents the mean value of the health status related to the battery parameters, Represents the variance of the health state associated with the battery parameters.

[0077] Furthermore, for the current training battery parameters, the prior distribution of each training battery parameter, the first likelihood function and the current training battery parameters can be combined to determine the first posterior distribution corresponding to the predicted health state. Specifically, for each training battery parameter, the Bayesian theorem can be used to combine the prior distribution of each training battery parameter, the first likelihood function corresponding to the health state and the current training battery parameters to obtain the first posterior distribution P(θ i |SOH, current training battery parameters)∝P(SOH|θ1,θ2,...,θ n )·P(θ i ), where P(θ i |SOH, current training battery parameters) represents the probability distribution of the health state being SOH when the battery parameters are given.

[0078] Parameter information can also be extracted from the first posterior distribution to update the target node corresponding to the health state in the Bayesian network. Specifically, the parameters of the target node corresponding to the health state can be updated by calculating the expected value, median, etc. of the first posterior distribution.

[0079] Of course, after updating the Bayesian network based on the current training battery parameters, the first posterior distribution can be used as a new prior distribution, and the above steps can be repeated based on other training battery parameters to achieve iterative parameter updates for the target nodes corresponding to the health status in the Bayesian network.

[0080] The input of the Bayesian network is the current training battery parameters contained in the current training battery data and the actual health state corresponding to the current battery data, and the output is the posterior probability distribution of the health state. The posterior probability distribution of the predicted health state reflects the probability distribution of the battery health state when the battery data of the battery to be tested is the current training battery data. Specifically, it is the probability of the battery to be tested being in different health states given the current training battery data, indicating the possibility of the battery to be tested being in different health states.

[0081] In each iteration, for each battery parameter node, its posterior distribution parameters are calculated based on the current training battery parameters contained in the current training battery data and the actual health state corresponding to the current battery data. This process can update the specific values ​​of the battery parameters to better reflect the current training battery parameters contained in the current training battery data and the actual health state corresponding to the current battery data.

[0082] By continuously performing the above process, the parameters in the Bayesian network are continuously updated by the training battery parameters contained in each training battery data and the actual health status corresponding to each training battery data, so that the Bayesian network can better adapt to the actual data distribution.

[0083] In the embodiment of the present invention, the parameter update of the target node corresponding to the health state in the Bayesian network is achieved based on the parameters of each training battery.

[0084] Step 240, define a second likelihood function for the model parameters of the prediction model; for each of the training battery parameters, determine the second posterior distribution corresponding to each of the model parameters in combination with the prior distribution of each of the training battery parameters, the second likelihood function, the current training battery parameters and the current health status; and update the Bayesian network according to the second posterior distribution.

[0085] Specifically, based on the formula After determining the prior distribution of each training battery parameter, a second likelihood function can be defined for the model parameters of the prediction model To indicate that under given battery parameters, the model parameters are The probability of . Represents the model parameters of the prediction model. Since the prediction model includes multiple model parameters, represents the set of model parameters of the prediction model, represents the mean of the model parameters related to the battery parameters, represents the variance of the model parameters related to the battery parameters.

[0086] Furthermore, for the current training battery parameters, the prior distribution of each training battery parameter, the second likelihood function and the current training battery parameters can be combined to determine the second posterior distribution corresponding to the model parameters of the prediction model. Specifically, for each training battery parameter, the Bayesian theorem can be used to combine the prior distribution of each training battery parameter, the second likelihood function corresponding to the model parameters of the prediction model and the current training battery parameters to obtain the second posterior distribution corresponding to the model parameters of the prediction model. Current training battery parameters) in, Current training battery parameters) means that when the battery parameters are given, the model parameters are The probability distribution of .

[0087] Parameter information can also be extracted from the second posterior distribution to update the target nodes corresponding to the model parameters of the prediction model in the Bayesian network. Specifically, the parameters of the target nodes corresponding to the model parameters of the prediction model can be updated by calculating the expected value, median, etc. of the second posterior distribution.

[0088] Of course, after updating the Bayesian network based on the current training battery parameters, the second posterior distribution can be used as a new prior distribution, and the above steps can be repeated based on other training battery parameters to achieve iterative parameter updates of the target nodes corresponding to the model parameters of the prediction model in the Bayesian network.

[0089] In the embodiment of the present invention, the parameter update of the target node corresponding to the model parameter of the prediction model in the Bayesian network is realized based on the parameters of each training battery.

[0090] It should be noted that after updating the Bayesian network, the adjusted Bayesian network can be verified and tested. Specifically, appropriate performance indicators can be selected, such as root mean square error (RMSE), mean absolute error (MAE), etc., to quantify the difference between the prediction of the Bayesian network and the actual observation.

[0091] Step 250: When it is determined that any of the training battery parameters does not conform to the prior distribution of the training battery parameters, adjust the training battery parameters based on the prior distribution of the training battery parameters.

[0092] Specifically, after determining the prior distribution of each training battery parameter, if any training battery parameter in any training battery data does not conform to the prior distribution of the training battery parameter, the training battery parameter can be adjusted based on the prior distribution of the training battery parameter to achieve adjustment of the training battery parameter and reduce data offset of the training battery data. This can provide better quality training data for the training of the prediction model and improve the accuracy of the model.

[0093] In the embodiment of the present invention, it is possible to determine whether each training battery parameter in each training battery data is offset according to the prior distribution of the training battery parameters, and timely adjust the training battery data with data offset to provide training data with better quality for training the prediction model.

[0094] Step 260: Input each of the training battery data and the actual health status corresponding to each of the training battery data as training data into the neural network model, train the neural network model, and update the model parameters of the neural network model based on the Bayesian network to obtain the prediction model.

[0095] In one implementation, step 260 may specifically include:

[0096] The training battery data and the true health status corresponding to each training battery data are input into the neural network model as training data to train the neural network model; the model parameters of the neural network model are updated according to the training data and the second posterior distribution corresponding to each model parameter to obtain the prediction model.

[0097] The neural network model can be a recursive neural network RNN. For example, a long short-term memory network LSTM can be selected as the basic unit of the RNN. LSTM has a memory unit and a gating mechanism, which can better capture the time dependency of the battery health state. The neural network model is used to process the temporal nature of the battery health state. The neural network model can accept the training battery data and the real health state corresponding to each training battery data as input to learn the evolution trend of the battery health state.

[0098] Figure 3 A schematic diagram of a neural network model in another method for determining the health status of a battery provided by an embodiment of the present invention, such as Figure 3 As shown, after setting the initialization network parameters, number of iterations and other hyperparameters for the neural network model, the calculation method for determining the neural network model is as follows:

[0099] O t =g(V·S t )

[0100] S t =f(U·X t +W·S t-1 )

[0101] Among them, X represents the value of the input layer, S represents the value of the hidden layer, O represents the value of the output layer, U represents the weight matrix from the input layer to the hidden layer, V represents the weight matrix from the hidden layer to the output layer, and W represents the weight of the previous value of the hidden layer as the input this time.

[0102] Specifically, each training battery data and the actual health status corresponding to each training battery data are input into the neural network model as training data, and the neural network model is trained. During the training of the neural network model, the model parameters of the neural network model can be updated according to the training data and the second posterior distribution corresponding to each model parameter to obtain a prediction model.

[0103] In practical applications, the model parameters of the neural network model can be updated through optimization algorithms such as gradient descent, so that the obtained prediction model can accurately predict the health status of the battery to be tested.

[0104] After training the neural network model to obtain the prediction model, the data of the test set can be used to evaluate the prediction ability of the prediction model. The test battery data contained in the test set can be input into the prediction model, and the prediction model can determine the predicted health state corresponding to the predicted battery data. By determining the difference between the actual health state and the predicted health state corresponding to the test battery data, the accuracy and stability of the prediction model can be evaluated.

[0105] Of course, you can also use the reserved test set to verify the performance of the prediction model and ensure that the model has good generalization performance for unseen data. Specifically, you can compare the prediction model with the BN model and the RNN model, and use the preset performance indicators to evaluate the performance of each model. For example, you can use the mean square error, root mean square error, mean absolute error, etc. to verify the accuracy of the prediction model in determining the battery health status.

[0106] In an embodiment of the present invention, model training of a neural network model is implemented through each training battery data and the actual health status corresponding to each training battery data and a Bayesian network, thereby obtaining a prediction model that can be used to determine the health status of a battery based on the battery data of the battery.

[0107] Step 270: Obtain battery data of the battery to be tested.

[0108] The battery data includes a plurality of battery parameters, and the battery parameters include at least voltage, current, temperature and cycle number.

[0109] Step 280: Input the battery data into a pre-trained prediction model, so that the prediction model analyzes the health status of the battery to be tested according to the battery data to determine the health status of the battery to be tested.

[0110] As mentioned above, the battery data of the battery to be tested is input into a pre-trained prediction model, and the prediction model can analyze the health status of the battery to be tested based on the battery data of the battery to be tested to obtain the SOH of the battery to be tested. The prediction model obtained by Bayesian network training can analyze the SOH of the battery to be tested based on the battery data of the battery to be tested, so that the battery data can be processed more comprehensively and the SOH of the battery to be tested can be determined with higher accuracy.

[0111] In an embodiment of the present invention, after obtaining the battery data of the battery to be tested, the battery data of the battery to be tested can be input into a pre-trained prediction model. The prediction model can analyze the health status of the battery to be tested based on the battery data of the battery to be tested, thereby determining a more accurate health status of the battery to be tested.

[0112] The method for determining the health status of a battery provided in an embodiment of the present invention includes: obtaining training battery data and the actual health status corresponding to each of the training battery data; taking each of the training battery parameters as a node, the predicted health status corresponding to the training battery data and the model parameters of the prediction model as the target node, and constructing a Bayesian network; determining the prior distribution of each of the training battery parameters, and defining a first likelihood function for the predicted health status; for each of the training battery parameters, combining the prior distribution of each of the training battery parameters, the first likelihood function and the current training battery parameters, determining the first posterior distribution corresponding to the predicted health status; updating the Bayesian network according to the first posterior distribution; defining a second likelihood function for the model parameters of the prediction model; for each of the training battery parameters, combining the prior distribution of each of the training battery parameters, the second likelihood function and the current training battery parameters number, current training battery parameters and current health status, determine the second posterior distribution corresponding to each of the model parameters; update the Bayesian network according to the second posterior distribution; determine that when any of the training battery parameters does not conform to the prior distribution of the training battery parameters, adjust the training battery parameters based on the prior distribution of the training battery parameters; input each of the training battery data and the true health status corresponding to each of the training battery data as training data into a neural network model, train the neural network model, and update the model parameters of the neural network model based on the Bayesian network to obtain the prediction model; obtain battery data of the battery to be tested; input the battery data into a pre-trained prediction model so that the prediction model analyzes the health status of the battery to be tested according to the battery data and determines the health status of the battery to be tested.The above technical scheme can first obtain the training battery data and the actual health status corresponding to each training battery data. After constructing a Bayesian network with each training battery parameter as a node, the predicted health status corresponding to the training battery data and the model parameters of the prediction model as the target node, the prior distribution of each training battery parameter can be determined, and a first likelihood function can be defined for the predicted health status; for each training battery parameter, the first posterior distribution corresponding to the predicted health status is determined in combination with the prior distribution of each training battery parameter, the first likelihood function and the current training battery parameter; the Bayesian network is updated according to the first posterior distribution to achieve parameter update of the target node corresponding to the health status in the Bayesian network, and a second likelihood function can also be defined for the model parameters of the prediction model; for each training battery parameter, the first posterior distribution corresponding to each model parameter is determined in combination with the prior distribution of each training battery parameter, the second likelihood function, the current training battery parameter and the current health status. two posterior distributions; updating the Bayesian network according to the second posterior distribution to update the parameters of the target nodes corresponding to the model parameters of the prediction model in the Bayesian network; and determining whether the parameters of each training battery in each training battery data are offset according to the prior distribution of the training battery parameters, and timely adjusting the training battery data with data offset to provide good quality training data for the training of the prediction model; and then training the neural network model through the training battery data and the real health status corresponding to each training battery data and the Bayesian network to obtain a prediction model that can be used to determine the health status of the battery according to the battery data of the battery; after obtaining the battery data of the battery to be tested, the battery data of the battery to be tested can be input into the pre-trained prediction model, and the prediction model can analyze the health status of the battery to be tested according to the battery data of the battery to be tested, so as to determine a more accurate health status of the battery to be tested.

[0113] Figure 4 This is a schematic diagram of the structure of a battery health status determination device provided by an embodiment of the present invention, which can be applied to situations where the battery health status needs to be accurately determined. The device can be implemented by software and / or hardware and is generally integrated in a computer device.

[0114] like Figure 4 As shown, the device comprises:

[0115] An acquisition module 410 is used to acquire battery data of a battery to be tested, wherein the battery data includes a plurality of battery parameters;

[0116] The analysis module 420 is used to input the battery data into a pre-trained prediction model, so that the prediction model analyzes the health state of the battery to be tested according to the battery data to determine the health state of the battery to be tested.

[0117] The device for determining the health status of a battery provided in this embodiment obtains battery data of a battery to be tested, wherein the battery data includes a plurality of battery parameters; the battery data is input into a pre-trained prediction model, so that the prediction model analyzes the health status of the battery to be tested according to the battery data, and determines the health status of the battery to be tested. In the above technical scheme, after obtaining the battery data of the battery to be tested, the battery data of the battery to be tested can be input into a pre-trained prediction model, and the prediction model can analyze the health status of the battery to be tested according to the battery data of the battery to be tested. Since the prediction model is obtained by Bayesian network training, when the prediction model analyzes the health status of the battery to be tested according to the battery data of the battery to be tested, the battery data of the battery to be tested can be analyzed more comprehensively, so as to determine the health status of the battery to be tested with higher accuracy.

[0118] In one embodiment, the battery parameters include at least voltage, current, temperature and cycle number.

[0119] Based on the above embodiment, the device further includes:

[0120] A training module is used to obtain training battery data and the real health status corresponding to each training battery data, wherein the training battery data includes multiple training battery parameters; each training battery parameter is used as a node, and the predicted health status corresponding to the training battery data and the model parameters of the prediction model are used as target nodes to construct a Bayesian network; each training battery data and the real health status corresponding to each training battery data are input into a neural network model as training data, the neural network model is trained, and the model parameters of the neural network model are updated based on the Bayesian network to obtain the prediction model.

[0121] Based on the above embodiment, the training module is further used for:

[0122] After constructing a Bayesian network by taking each of the training battery parameters as a node, the predicted health state corresponding to the training battery data and the model parameters of the prediction model as target nodes, the prior distribution of each of the training battery parameters is determined, and a first likelihood function is defined for the predicted health state; for each of the training battery parameters, a first posterior distribution corresponding to the predicted health state is determined in combination with the prior distribution of each of the training battery parameters, the first likelihood function and the current training battery parameters; and the Bayesian network is updated according to the first posterior distribution.

[0123] Based on the above embodiment, the training module is further used for:

[0124] After constructing a Bayesian network by taking each of the training battery parameters as a node, the predicted health status corresponding to the training battery data and the model parameters of the prediction model as target nodes, a second likelihood function is defined for the model parameters of the prediction model; for each of the training battery parameters, a second posterior distribution corresponding to each of the model parameters is determined in combination with the prior distribution of each of the training battery parameters, the second likelihood function, the current training battery parameters and the current health status; and the Bayesian network is updated according to the second posterior distribution.

[0125] In one implementation, each of the training battery data and the real health status corresponding to each of the training battery data are input as training data into a neural network model, the neural network model is trained, and the model parameters of the neural network model are updated based on the Bayesian network to obtain the prediction model, including:

[0126] The training battery data and the true health status corresponding to each training battery data are input into the neural network model as training data to train the neural network model; the model parameters of the neural network model are updated according to the training data and the second posterior distribution corresponding to each model parameter to obtain the prediction model.

[0127] Based on the above embodiment, the device further includes:

[0128] The adjustment module is used to adjust the training battery parameters based on the prior distribution of the training battery parameters when it is determined that any of the training battery parameters does not conform to the prior distribution of the training battery parameters.

[0129] The battery health status determination device provided in the embodiment of the present invention can execute the battery health status determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the battery health status determination method.

[0130] It is worth noting that in the embodiment of the above-mentioned battery health status determination device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0131] Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 5 A block diagram of an exemplary computer device 5 suitable for use in implementing embodiments of the present invention is shown. Figure 5 The computer device 5 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0132] like Figure 5 As shown, the computer device 5 is in the form of a general-purpose computing electronic device. The components of the computer device 5 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).

[0133] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0134] The computer device 5 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 5, including volatile and non-volatile media, removable and non-removable media.

[0135] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 5 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0136] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0137] The computer device 5 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), one or more devices that enable a user to interact with the computer device 5, and / or any device that enables the computer device 5 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 22. Furthermore, the computer device 5 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 20. Figure 5 As shown, the network adapter 20 communicates with other modules of the computer device 5 via the bus 18. It should be understood that although Figure 5 Not shown, other hardware and / or software modules may be used in conjunction with computer device 5, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0138] The processing unit 16 executes various functional applications and page displays by running the program stored in the system memory 28, for example, implementing the battery health status determination method provided in the embodiment of the present invention, the method comprising:

[0139] Acquire battery data of a battery to be tested, wherein the battery data includes a plurality of battery parameters;

[0140] The battery data is input into a pre-trained prediction model, so that the prediction model analyzes the health state of the battery to be tested according to the battery data to determine the health state of the battery to be tested.

[0141] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the method for determining the health status of a battery provided in any embodiment of the present invention.

[0142] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, a method for determining a health state of a battery provided in an embodiment of the present invention is implemented, for example, the method comprising:

[0143] Acquire battery data of a battery to be tested, wherein the battery data includes a plurality of battery parameters;

[0144] The battery data is input into a pre-trained prediction model, so that the prediction model analyzes the health state of the battery to be tested according to the battery data to determine the health state of the battery to be tested.

[0145] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0146] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0147] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0148] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0149] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0150] In addition, the acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations.

[0151] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for determining a battery health status, characterized in that: include: Acquire battery data of a battery to be tested, wherein the battery data includes a plurality of battery parameters; Inputting the battery data into a pre-trained prediction model, so that the prediction model analyzes the health state of the battery to be tested according to the battery data to determine the health state of the battery to be tested; The method for determining the health status of a battery is characterized in that the training process of the prediction model includes: Acquire training battery data and the real health status corresponding to each training battery data, wherein the training battery data includes a plurality of training battery parameters; The training battery parameters are used as nodes, and the predicted health status corresponding to the training battery data and the model parameters of the prediction model are used as target nodes to construct a Bayesian network; The training battery data and the real health status corresponding to the training battery data are input into the neural network model as training data, the neural network model is trained, and the model parameters of the neural network model are updated based on the Bayesian network to obtain the prediction model.

2. The method for determining the health status of a battery according to claim 1, characterized in that: The battery parameters include at least voltage, current, temperature and cycle number.

3. The method for determining the health status of a battery according to claim 1, characterized in that: After constructing a Bayesian network by taking each of the training battery parameters as a node, the predicted health state corresponding to the training battery data and the model parameters of the prediction model as the target node, the method further includes: determining a prior distribution of each of the training battery parameters to define a first likelihood function for the predicted health state; For each of the training battery parameters, combining the prior distribution of each of the training battery parameters, the first likelihood function and the current training battery parameters, determining a first posterior distribution corresponding to the predicted health state; The Bayesian network is updated according to the first posterior distribution.

4. The method for determining the health status of a battery according to claim 3, characterized in that: After constructing a Bayesian network by taking each of the training battery parameters as a node, the predicted health state corresponding to the training battery data and the model parameters of the prediction model as the target node, the method further includes: defining a second likelihood function for model parameters of the prediction model; For each of the training battery parameters, determine a second posterior distribution corresponding to each of the model parameters in combination with the prior distribution of each of the training battery parameters, the second likelihood function, the current training battery parameters and the current health status; The Bayesian network is updated according to the second posterior distribution.

5. The method for determining the health status of a battery according to claim 4, characterized in that: Inputting each of the training battery data and the real health status corresponding to each of the training battery data as training data into a neural network model, training the neural network model, and updating the model parameters of the neural network model based on the Bayesian network to obtain the prediction model, including: Inputting each of the training battery data and the real health status corresponding to each of the training battery data as training data into a neural network model to train the neural network model; The model parameters of the neural network model are updated according to the training data and the second posterior distribution corresponding to each of the model parameters to obtain the prediction model.

6. The method for determining the health status of a battery according to claim 3, characterized in that: Before inputting each of the training battery data and the real health status corresponding to each of the training battery data as training data into the neural network model, the method further includes: When it is determined that any of the training battery parameters does not conform to the prior distribution of the training battery parameters, the training battery parameters are adjusted based on the prior distribution of the training battery parameters.

7. A device for determining the health status of a battery, characterized in that: include: An acquisition module, used to acquire battery data of a battery to be tested, wherein the battery data includes a plurality of battery parameters; An analysis module, used for inputting the battery data into a pre-trained prediction model, so that the prediction model analyzes the health state of the battery to be tested according to the battery data to determine the health state of the battery to be tested; A training module is used to obtain training battery data and the real health status corresponding to each training battery data, wherein the training battery data includes multiple training battery parameters; each training battery parameter is used as a node, and the predicted health status corresponding to the training battery data and the model parameters of the prediction model are used as target nodes to construct a Bayesian network; each training battery data and the real health status corresponding to each training battery data are input into a neural network model as training data, the neural network model is trained, and the model parameters of the neural network model are updated based on the Bayesian network to obtain the prediction model.

8. A computer device, characterized in that: The computer device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the battery health status determination method as described in any one of claims 1-6.

9. A storage medium containing computer executable instructions, characterized in that: When the computer executable instructions are executed by a computer processor, they are used to perform the method for determining the health status of a battery as described in any one of claims 1 to 6.