Storage battery health state assessment method, system and device and storage medium

Abstract

The invention relates to a storage battery health state assessment method, system and device and a storage medium. The method comprises the following steps: S1, monitoring the working state of a storage battery; s2, when the voltage of the storage battery is suddenly dropped and regenerated, acquiring corresponding preset characteristic parameters of the storage battery when the voltage of the storage battery is suddenly dropped; and S3, according to the preset characteristic parameters and based on a preset neural network algorithm, constructing a health state evaluation model corresponding to the storage battery, so as to perform health state evaluation of the to-be-tested storage battery through the health state evaluation model. According to the invention, the method can achieve the accurate health state evaluation of the lithium ion storage battery, and guarantees the safe and stable operation of the lithium ion storage battery.

Description

Technical Field

[0001] The present invention relates to the field of battery technology, and more specifically, to a battery health status assessment method, system, device and storage medium. Background Art

[0002] The application scope of lithium-ion batteries has become more and more extensive. However, some problems in the use of lithium-ion batteries are also affecting the development of industries such as electricity and communications. For example, harsh computer room environment, poor power supply conditions, many lithium-ion battery discharges, deep discharge, over-discharge, lithium-ion battery aging, and high ambient temperature can cause lithium-ion battery damage or a sharp drop in lithium-ion battery capacity. Even if only one lithium-ion battery deteriorates, it will affect the performance of the entire group of lithium-ion batteries, thereby affecting the normal power supply of the equipment. Therefore, timely and reliable monitoring and health status assessment of lithium-ion batteries are of great significance for maintaining the normal operation of communication system equipment. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a battery health status assessment method, system, device and storage medium for the above-mentioned technical defects of the prior art.

[0004] The technical solution adopted by the present invention to solve the technical problem is: construct a battery health status assessment method, the method comprising the following steps:

[0005] S1. Monitor the working status of the battery;

[0006] S2. When a voltage drop regeneration phenomenon occurs in the battery, obtaining preset characteristic parameters corresponding to the battery when the voltage drops suddenly;

[0007] S3. Construct a health status assessment model corresponding to the battery according to the preset characteristic parameters and based on a preset neural network algorithm, so as to perform health status assessment on the battery to be tested through the health status assessment model and obtain a health status assessment result of the battery according to the health status assessment model.

[0008] Preferably, in an embodiment of the battery health status assessment method of the present invention, in step S1, monitoring the working status of the battery includes:

[0009] Monitor the corresponding floating charge voltage information, charge and discharge current information and temperature information during the operation of the battery;

[0010] Whether the battery has a voltage drop recovery phenomenon is determined based on the floating charge voltage information, the charge and discharge current information, and the temperature information.

[0011] Preferably, in an embodiment of the battery health status assessment method of the present invention, in the step S2, the obtaining of preset characteristic parameters corresponding to the battery when the voltage drops sharply comprises:

[0012] The valley voltage data, valley current data, peak voltage data, peak current data, voltage difference data, current difference data and temperature data corresponding to the battery when the voltage drops sharply are obtained.

[0013] Preferably, in an embodiment of the battery health status assessment method of the present invention, in step S3, the preset neural network algorithm includes a BP neural network algorithm, and the health status assessment model corresponding to the battery is constructed according to the preset characteristic parameters and based on the preset neural network algorithm, including:

[0014] Performing dimensionality reduction processing on the preset characteristic parameters according to a linear discriminant algorithm to obtain characteristic parameters after dimensionality reduction;

[0015] The Pearson correlation coefficient is used to determine the correlation between the reduced-dimensional feature parameters and the battery health status, and the feature parameters whose correlation reaches a preset value among the reduced-dimensional feature parameters are obtained as health feature parameters;

[0016] The health status assessment model is established according to the health characteristic parameters and the BP neural network algorithm.

[0017] Preferably, in an embodiment of the battery health status assessment method of the present invention, the health status assessment model is established according to the health feature parameters and the BP neural network algorithm, including:

[0018] Normalizing the health characteristic parameters to obtain a corresponding sample set, and dividing the sample set according to a preset ratio to obtain a training set and a test set;

[0019] An evaluation network is constructed using a three-layer BP neural network as an evaluation algorithm and a tan·sig function as an activation function, wherein the input layer of the evaluation network includes a first number of neurons, the output layer includes a second number of neurons, and the hidden layer includes a third number of neurons;

[0020] The evaluation network is trained according to the training set, and the evaluation network is tested and analyzed based on the test set to obtain a final health status evaluation model of the evaluation network with the best output result.

[0021] Preferably, in an embodiment of the battery health status assessment method described in the present invention, the first number is 7, the second number is 1, and the third number is 10.

[0022] Preferably, in an embodiment of the battery health status assessment method of the present invention, the step of dividing the sample set according to a preset ratio to obtain a training set and a test set includes:

[0023] The sample set is divided into a training set and a test set in a ratio of 7:3.

[0024] The present invention also constructs a battery health status assessment system, including: a module for executing the method as described above.

[0025] The present invention also constructs a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the battery health status assessment method as described above is implemented.

[0026] The present invention also constructs an electronic device, comprising a memory and a processor;

[0027] The memory is used to store computer programs;

[0028] The processor is used to execute the computer program to implement the battery health status assessment method as described above.

[0029] The battery health status assessment method, system, device and storage medium implementing the present invention have the following beneficial effects: accurate health status assessment of lithium-ion batteries can be achieved, ensuring safe and stable operation of lithium-ion batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0031] Figure 1 It is a program flow chart of an embodiment of a method for evaluating the health status of a battery according to the present invention;

[0032] Figure 2 is a flowchart of another embodiment of a method for evaluating the health status of a battery according to the present invention;

[0033] Figure 3 is a flowchart of another embodiment of a method for evaluating the health status of a battery according to the present invention;

[0034] Figure 4 It is a curve diagram showing the phenomenon of sudden voltage drop and regeneration in a battery. DETAILED DESCRIPTION

[0035] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0036] like Figure 1As shown in FIG. 1 , an embodiment of a method for evaluating the health status of a battery according to the present invention is provided. Figure 1 In one embodiment of a battery health status assessment method of the present invention, the following steps are included: S1, monitoring the working status of the battery; S2, when the battery has a voltage drop regeneration phenomenon, obtaining the preset characteristic parameters corresponding to the battery when the voltage drops sharply; S3, constructing a health status assessment model corresponding to the battery according to the preset characteristic parameters and based on a preset neural network algorithm, so as to evaluate the health status of the battery to be tested through the health status assessment model.

[0037] Based on the above step S1, during the operation of the battery, the working state of the battery can be monitored by a corresponding monitoring circuit or device. In the specific monitoring process, it can be understood as monitoring the working parameters of the battery when it is working. In a specific embodiment, the corresponding positions of the battery are monitored by various sensors to collect corresponding parameters. For example, the voltage signal of the battery is obtained by a voltage sensor, the current signal of the battery is obtained by a current sensor, and the temperature signal of the battery is obtained by a temperature sensor. The various collected signals obtained are amplified and filtered by a signal conditioning module to obtain a collected signal with quality that meets the requirements. The final signal is converted into analog to digital to obtain the corresponding collected data.

[0038] Based on the above step S2, during the process of monitoring the working state of the battery, when the battery voltage drops suddenly and regenerates, the corresponding preset characteristic parameters are obtained according to the monitoring parameters of the battery when the voltage drops suddenly. The preset characteristic parameters can be understood as the required characteristic quantities selected from a large number of monitoring parameters.

[0039] Based on the above step S3, after obtaining the preset characteristic parameters, a preset neural network algorithm can be selected, and a health status assessment model can be constructed according to the preset neural network algorithm. The model can be understood as a neural network model. Through the neural network model, the health status of the battery to be tested can be evaluated to obtain the corresponding evaluation results. It can be understood that the health status assessment model is constructed based on the sample battery. After obtaining the final health status assessment model, the health status of the battery to be tested can be evaluated.

[0040] Optional, such as Figure 2As shown, in the step S1, the monitoring of the working state of the battery includes: S11, monitoring the corresponding floating charge voltage information, charge and discharge current information and temperature information during the working process of the battery; S12, judging whether the battery has a voltage drop regeneration phenomenon according to the floating charge voltage information, charge and discharge current information and temperature information. Specifically, during the working process of the battery, it can be judged whether the battery has a voltage drop regeneration phenomenon by obtaining the floating charge voltage information, charge and discharge current information and temperature information corresponding to the battery. In this process, the obtained data can be processed and displayed by the host computer to judge whether the battery has a voltage drop regeneration phenomenon according to the displayed information. After the specifications of the battery have been set, the corresponding floating charge voltage is generally a pre-set fixed value, and the cut-off current of the floating charge stage is also specified to be less than a fixed value. When the above conditions are met, it indicates that the battery is fully charged, and the temperature information will also be recorded by the battery management system. When the battery enters the discharge state from the fully charged state, the battery will experience a voltage drop and rise phenomenon.

[0041] Optionally, in step S2, the obtaining of preset characteristic parameters corresponding to the battery when the voltage drops suddenly includes: obtaining valley voltage data, valley current data, peak voltage data, peak current data, voltage difference data, current difference data and temperature data corresponding to the battery when the voltage drops suddenly. Specifically, when a sudden voltage drop is detected, the valley voltage data, valley current data, peak voltage data, peak current data, voltage difference data, current difference data and temperature data corresponding to the sudden voltage drop can be obtained, and the data can be used as preset characteristic parameters. Figure 4 As shown, it is a schematic diagram of a battery steep drop recovery curve. In the curve schematic diagram, the current data recorded at the valley voltage moment is the valley current data, and the current data recorded at the platform voltage moment is the peak current data. Among them, the voltage difference is the difference between the platform voltage and the valley voltage in the above-mentioned steep drop recovery curve diagram, and the current difference is the difference between the current at the corresponding moment of the platform voltage and the current at the corresponding moment of the valley voltage. Because the health state estimation method is oriented to the battery, most of them work in the floating charge state and rarely discharge, so in this embodiment, the floating charge life is used to represent the health state of the battery. Considering that the valley voltage and platform voltage of the steep drop and rise phenomenon have a linear relationship with the capacity, the health state of the battery can be evaluated with the help of the "steep drop and rise" phenomenon of the battery. And extracting indirect health indicators such as the voltage and current difference from the voltage, current and temperature data of the battery at this stage can comprehensively analyze the existing data, so that the health state of the battery is estimated more accurately.

[0042] Optional, such as Figure 3As shown, in the step S3, the preset neural network algorithm includes a BP neural network algorithm, wherein the BP neural network algorithm is a back propagation neural network (Back Propagation neural network), and the health status assessment model corresponding to the battery is constructed according to the preset characteristic parameters and based on the preset neural network algorithm, including: S31, reducing the dimension of the preset characteristic parameters according to a linear discriminant algorithm to obtain the reduced-dimensional characteristic parameters; S32, using the Pearson correlation coefficient to judge the correlation between the reduced-dimensional characteristic parameters and the battery health status, and obtaining the characteristic parameters whose correlation reaches the preset value in the reduced-dimensional characteristic parameters as health characteristic parameters; S33, establishing the health status assessment model according to the health characteristic parameters and the BP neural network algorithm.

[0043] Specifically, based on step S31, a set of battery data sets is selected:

[0044] D={(x 1,n1 ,y1),(x 2,n2 ,y2),(x 3,n3 ,y3),…,(x m,nm ,y4)}

[0045] where x i,ni is a ni×7 matrix, used to represent the battery health status as y i All feature samples corresponding to the battery obtained at the time of the battery health state are all data corresponding to the preset feature parameters of the battery. Among them, i is 1, 2, 3, ... m, ni represents the number of feature samples corresponding to each type of preset feature parameters in the current battery health state, and the number 7 represents that the number of categories of the preset feature parameters is 7. In some embodiments, the number of categories of the preset feature parameters may not be 7. i Corresponding to different battery health states, m represents the total number of battery health state categories. In one embodiment, the battery health state can be divided according to actual application scenarios to obtain different total number of categories.

[0046] Then calculate the intra-class scatter matrix and the inter-class scatter matrix. The specific process can be: first define D i Indicates that the battery health status is y i When , all sets of characteristic samples corresponding to the preset characteristic parameters of the battery are x i,ni Then we get the battery health status as y i When , the sample mean vector corresponding to the preset characteristic parameters of the battery is

[0047]

[0048] Among them, the battery health status is y iWhen , the sample mean vector μ i is a 1×7 vector. Based on the above process, a total of m 1×7 mean matrices can be obtained to correspond to the sample mean vectors under different battery health states.

[0049] Furthermore, the battery health status is obtained as y i The calculation formula of the feature class distribution matrix is ​​as follows:

[0050]

[0051] Corresponding to the scatter matrix formula within the total category

[0052]

[0053] Get the dispersion degree within the feature sample. w It is a 7×7 matrix, which is used to reflect the internal dispersion of samples corresponding to each type of preset characteristic parameters corresponding to batteries in the same battery health state.

[0054] The overall mean vector of the characteristic samples corresponding to all batteries in all battery health states can be obtained by the formula

[0055]

[0056] Calculated. Where N represents the total number of samples corresponding to the battery, N = n1 + n2 + n3 + ... + nm. Then calculate the dispersion degree between each feature, which is specifically obtained according to the following inter-class spread matrix calculation formula:

[0057]

[0058] Among them, S B is a 7×7 matrix, reflecting the dispersion of the corresponding features of the battery under different battery health conditions. w ―1 S B , solve the eigenvalue of the matrix, get 7 eigenvalues ​​and corresponding 7×1 eigenvectors, arrange the 7 eigenvalues ​​in order, the larger the eigenvalue, the more important its corresponding eigenvector is, take the first two eigenvalues ​​and their corresponding eigenvectors to form a 7×2 projection matrix W, that is, reduce the original feature data to two dimensions, then the projected data points are given by the formula X=WX original Calculated. Among them, X original is the original feature data, with a size of N×7. In particular, the original feature data consists of the features corresponding to the health status of all categories in the data set D. X is the projected data, with a data size of N×2, completing all the dimensionality reduction work. It should be noted that the dimensionality reduction of the seven feature quantities is performed simultaneously here.

[0059] Based on step S32, the correlation between the preset feature parameters and the battery health status is analyzed by the Pearson (person) correlation coefficient method, and the health features with strong correlation with the battery health status are selected according to the correlation analysis results. Its value range is [-1, 1]; that is, it can be understood that the correlation coefficient method is used to analyze the correlation between the feature input and the battery health status, and the value range is [-1, 1]. Generally, 0-0.2 indicates extremely weak correlation or no correlation, 0.2-0.4 indicates weak correlation, 0.4-0.6 indicates moderate correlation, 0.6-0.8 indicates strong correlation, and 0.8-1 indicates extremely strong correlation. A positive value indicates a positive correlation, and a negative value indicates a negative correlation. Among them, the expression of the Pearson correlation coefficient is:

[0060]

[0061] In the correlation coefficient expression, X is the battery feature input after dimensionality reduction, with a size of N×2, and Y is the corresponding battery health state set, with a size of N×1. In particular, here Y is the set of all health states y in the data set D. i Cov(X,Y) is used to solve the covariance between the battery characteristic parameters and the battery health status after dimension reduction, σ X represents the standard deviation of the input features after dimensionality reduction, σ Y Indicates the standard deviation of the battery health status. μ X Describes the overall mean of the input features after dimensionality reduction, μ Y Describes the overall mean of the battery's state of health.

[0062]

[0063] Y=[Y1 Y2 Y3…Y N ] T

[0064] Among them, X1 represents the first dimension feature after dimensionality reduction, and X2 represents the second dimension feature after dimensionality reduction. 1i , X 2i , Y i ) constitutes a sample set (1≤i≤N), X 1i represents the i-th feature in the first dimension after dimensionality reduction, X 2i represents the i-th feature in the second feature after dimensionality reduction, Y i Indicates the corresponding i-th battery health status, and y i The same. Based on the sample set, the overall mean and standard deviation are as follows:

[0065]

[0066] In particular, in the above formula, E(X1) represents the overall mean of the first dimension characteristics, E(X2) represents the overall mean of the second dimension characteristics, and E(Y) represents the overall mean of the health status. is the standard deviation of the first dimension feature, is the standard deviation of the second dimension feature, σ Y is the standard deviation of the health status. N is the number of samples of the feature after dimensionality reduction.

[0067] According to the person coefficient method expression, the correlation coefficient between X1 and Y and the correlation coefficient between X2 and Y are calculated respectively:

[0068]

[0069]

[0070] In the above formula, Cov(X1,Y) represents the covariance between the first dimension feature and the health status, and Cov(X2,Y) represents the covariance between the second dimension feature and the health status. represents the correlation coefficient between the first dimension feature and the corresponding health status, Represents the correlation coefficient between the second dimension feature and the corresponding health status.

[0071] Specific data correlation analysis: Due to the limited data collected, the correlation is limited when combined with specific experimental data analysis. The above 7 features are selected based on the voltage, current and other curves collected under the current discharge condition, which show a trend of change related to the capacity. In order to reduce the computational complexity of the subsequent health assessment model and make the model output more accurate, the correlation analysis between input features and output is an important part of feature engineering, and dimensionality reduction does not change the correlation between feature input and output. The selected features are for the current working conditions. Once the working conditions change, the applicability of the features changes, and the correlation of the features needs to be re-analyzed.

[0072] Based on step S33, after obtaining the characteristic quantity optimized by the person correlation coefficient method, that is, the health characteristic parameter, the health status assessment model can be established according to the health characteristic parameter and based on the BP neural network algorithm.

[0073] Furthermore, the health status assessment model is established according to the health characteristic parameters and the BP neural network algorithm, including: S331, normalizing the health characteristic parameters to obtain a corresponding sample set, and dividing the sample set according to a preset ratio to obtain a training set and a test set; S332, using a three-layer BP neural network as an evaluation algorithm, and using a tan·sig function as an activation function to construct an evaluation network, wherein the input layer of the evaluation network includes a first number of neurons, the output layer includes a second number of neurons, and the hidden layer includes a third number of neurons; S333, training the evaluation network according to the training set, and testing and analyzing the evaluation network based on the test set to obtain a final health status assessment model of the evaluation network with the optimal output result.

[0074] That is, the evaluation network is constructed by using a three-layer BP neural network as an evaluation algorithm and a tan·sig function as an activation function. The three-layer structure includes an input layer, a hidden layer, and an output layer, and the number of neurons in the input layer is set to a first number, the number of neurons in the output layer is set to a second number, and the number of neurons in the hidden layer is set to a third number. In a specific embodiment, the input layer can be set to include 7 neurons, the output layer can include 1 neuron, and the hidden layer can include 10 neurons. The obtained health feature parameters are normalized to obtain data as a sample set. The sample is divided into a training set and a test set, and the evaluation network is trained and evaluated according to the training set and the test set, and the optimal evaluation network is obtained as a health status evaluation model. The process of training and evaluating the neural network based on the training set and the test set can be implemented using a relatively common neural network training method. For example, in the process of training the BP neural network model based on the training set, the weight is adjusted according to the output of the BP neural network to achieve the optimal output result. In the process of testing and analyzing the BP neural network model based on the test set, the health status output of the model is evaluated using the mean absolute error and the mean square error. A separate process description is not performed here. Finally, the optimal BP neural network model is selected for battery health status evaluation. In the above steps, the health features with strong correlation are selected through correlation analysis, because up to 7 features may be selected. The BP neural network will not change the propagation rules between layers due to the change in the number of input features, and the weights between layers will be dynamically adjusted due to the existence of the back propagation mechanism to make the BP network output closer to the expected value.

[0075] In one embodiment, the specific process of normalizing the health feature data can be: assuming that the number of health features after optimization is m and the number of samples is n, the unnormalized data can be expressed as:

[0076] X i =[x i,1 ,x i,2 ,x i,3,x i,j ,…,x i,n ]

[0077] Where 1≤i≤m, 1≤j≤n. The normalization calculation process is:

[0078] X i(归一化) =(x i,j –x i,j(最小值) ) / (x i,j(最大值) ―x i,j(最小值) ).

[0079] The valley voltage and valley current data extracted from the acquisition curve with a discharge current of 6.9A are used as an example to illustrate the normalized data process: Description of original data: A total of 15 points of valley voltage and valley current data are collected in the acquisition curve. Among them, valley voltage: X1 = [12.603, 12.663, 12.386, 12.413, 12.415, 12.422, 12.485, 12.511, 12.401, 12.421, 12.475, 12.422, 12.468, 12.496, 12.609]; valley current: X2 = [6.754, 6.772, 6.798, 6.783, 6.783, 6.779, 6.782, 6.792, 6.763, 6.761, 6.773, 6.779, 6.769, 6.763, 6.792]; normalized calculation formula: X 1,2(归一化) =(X 1,2 -X 1,2(最小值) ) / (X 1,2(最大值) -X 1,2(最小值) ); Normalized calculation results: Valley voltage normalized data: [0.7833935,1,0,0.09747292,0.10469314,0.1299639,0.35740072,0.45126354,0.05415162,0.12635379,0.32129964,0.1299639,0.29602888,0.39711191,0.80505415]. Valley current normalized data: [0, 0.40909091, 1, 0.65909091, 0.65909091, 0.56818182, 0.63636364, 0.86363636, 0.20454545, 0.15909091, 0.43181818, 0.56818182, 0.34090909, 0.20454545, 0.86363636].

[0080] In a specific embodiment, the sample set can be divided into a training set and a test set according to a ratio of 7:3. In an embodiment, the sample set can also be divided according to other ratios to ensure sufficient number of samples in the training set and the test set.

[0081] In addition, a battery health status assessment system of the present invention includes: a module for executing the method described above. Specifically, the specific coordination operation process between the modules of the battery health status assessment system here can refer to the above-mentioned battery health status assessment method, which will not be repeated here.

[0082] In addition, an electronic device of the present invention includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the battery health status assessment method as above. Specifically, according to an embodiment of the present invention, the process described with reference to the flowchart above can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed by an electronic device and when executed, it performs the above functions defined in the method of the embodiment of the present invention. The electronic device in the present invention can be a terminal such as a notebook, a desktop, a tablet computer, a smart phone, or a server.

[0083] In addition, a computer storage medium of the present invention stores a computer program thereon, and when the computer program is executed by a processor, the above battery health status assessment method is implemented. Specifically, it should be noted that the above-mentioned computer-readable medium of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may 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 of computer-readable storage media may include, but are not limited to: 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 the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a 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 computer readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0084] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0085] It can be understood that the above embodiments only express the preferred implementation modes of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the patent scope of the present invention. It should be pointed out that, for ordinary technicians in this field, the above technical features can be freely combined without departing from the concept of the present invention, and several deformations and improvements can be made, which all belong to the protection scope of the present invention. Therefore, all equivalent changes and modifications made to the scope of the claims of the present invention should belong to the coverage of the claims of the present invention.

Claims

1. A method for evaluating the health status of a battery, characterized in that: The method comprises the following steps: S1. Monitor the working status of the battery; S2. When a voltage drop regeneration phenomenon occurs in the battery, obtaining preset characteristic parameters corresponding to the battery when the voltage drops suddenly; S3. Construct a health status assessment model corresponding to the battery according to the preset characteristic parameters and based on a preset neural network algorithm, so as to perform health status assessment of the battery to be tested through the health status assessment model.

2. The battery health status assessment method according to claim 1, characterized in that: In the step S1, the monitoring of the working state of the battery includes: Monitor the corresponding floating charge voltage information, charge and discharge current information and temperature information during the operation of the battery; Whether the battery has a voltage drop recovery phenomenon is determined based on the floating charge voltage information, the charge and discharge current information, and the temperature information.

3. The battery health status assessment method according to claim 1, characterized in that: In the step S2, the step of obtaining the preset characteristic parameters corresponding to the battery when the voltage drops sharply includes: The valley voltage data, valley current data, peak voltage data, peak current data, voltage difference data, current difference data and temperature data corresponding to the battery when the voltage drops sharply are obtained.

4. The battery health status assessment method according to claim 3, characterized in that: In the step S3, the preset neural network algorithm includes a BP neural network algorithm, and the health status assessment model corresponding to the battery is constructed according to the preset characteristic parameters and based on the preset neural network algorithm, including: Performing dimensionality reduction processing on the preset characteristic parameters according to a linear discriminant algorithm to obtain characteristic parameters after dimensionality reduction; The Pearson correlation coefficient is used to determine the correlation between the reduced-dimensional feature parameters and the battery health status, and the feature parameters whose correlation reaches a preset value among the reduced-dimensional feature parameters are obtained as health feature parameters; The health status assessment model is established according to the health characteristic parameters and the BP neural network algorithm.

5. The battery health status assessment method according to claim 4, characterized in that: The health status assessment model is established according to the health characteristic parameters and the BP neural network algorithm, including: Normalizing the health characteristic parameters to obtain a corresponding sample set, and dividing the sample set according to a preset ratio to obtain a training set and a test set; An evaluation network is constructed using a three-layer BP neural network as an evaluation algorithm and a tan·sig function as an activation function, wherein the input layer of the evaluation network includes a first number of neurons, the output layer includes a second number of neurons, and the hidden layer includes a third number of neurons; The evaluation network is trained according to the training set, and the evaluation network is tested and analyzed based on the test set to obtain a final health status evaluation model of the evaluation network with the best output result.

6. The battery health status assessment method according to claim 5, characterized in that: The first number is 7, the second number is 1, and the third number is 10.

7. The battery health status assessment method according to claim 5, characterized in that: The step of dividing the sample set according to a preset ratio to obtain a training set and a test set includes: The sample set is divided into a training set and a test set in a ratio of 7:

3.

8. A battery health status assessment system, characterized in that: include: A module for executing the method according to any one of claims 1 to 7.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the battery health status assessment method as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is used to execute the computer program to implement the battery health status assessment method as described in any one of claims 1-7.

Images