Hybrid energy storage battery state monitoring method and system based on big data processing

Abstract

The present invention relates to a hybrid energy storage battery state monitoring method and system based on big data processing. According to the method, for the problems of the states of hybrid energy storage batteries being non-linear and being difficult to evaluate online, a charging sample information collection step, a discharging sample information collection step, a data collation and fusion step, a hybrid energy storage battery SOH estimation step, and a hybrid energy storage battery health level evaluation step are provided in sequence for achieving state monitoring of service hybrid energy storage batteries, wherein sample data information elements comprise voltage, current, power, temperature and internal resistance in the charging and discharging state, data is collated and fused, data prediction is carried out by means of a preset model, the states of the hybrid energy storage batteries can be covered in all directions, comprehensive evaluation is carried out, so that the evaluation accuracy is improved.

Description

A hybrid energy storage battery status monitoring method and system based on big data processing Technical Field

[0001] The present invention relates to the technical field of battery status prediction and health management, and in particular to a hybrid energy storage battery status monitoring method and system based on big data processing. Background Art

[0002] With the promotion of electric vehicles in China and the application of connected vehicle technology, more and more electric vehicles have entered the consumer market and are now collecting real-time driving data in accordance with the national standard (GBT32960). However, as the power source of electric vehicles, the battery capacity continues to decline with the increase in charge and discharge cycles and mileage. This reaction is a typical dynamic nonlinear electrochemical time-varying system. The internal parameters are difficult to measure in online applications, and the identification and estimation of its degradation state still pose significant challenges.

[0003] A battery's State of Health (SOH) refers to the ratio of the actual value of certain directly measurable or indirectly calculated performance parameters to the nominal value after a period of use under certain conditions. It is used to assess battery health and is generally expressed as a percentage. Furthermore, SOH is not only related to the battery's electrochemical system and manufacturing process, but also to the vehicle's driving conditions and the operating environment within the battery pack. For example, the aging of lead-acid batteries is affected by numerous factors, and battery aging experiments are limited by the full charge and discharge time and sample size. This makes the selection of a representative feature set based on a small sample size particularly important in battery SOH prediction.

[0004] This paper adopts a hybrid energy storage battery status monitoring method and system based on big data processing. Under the condition of relying on the long cycle of electric vehicle data collection, the implicit battery health status information and its evolution law are mined from the battery's rated information and status monitoring data (voltage, current, temperature, internal resistance, etc.) to realize battery SOH prediction.

[0005] Summary of the Invention

[0006] In order to overcome the shortcomings of the above-mentioned prior art, the present invention discloses a hybrid energy storage battery status monitoring method and system based on big data processing, which extracts the characteristic elements of five batteries under the charging and discharging states, and through sorting and fusion processing, performs SOH prediction through a preset model to effectively improve the accuracy of the prediction.

[0007] The technical solution adopted in this disclosure is:

[0008] A first aspect of an embodiment of the present invention provides a hybrid energy storage battery status monitoring method based on big data processing, which is applied to a hybrid energy storage battery status monitoring system for big data processing. The method includes:

[0009] Charging sample information collection step: obtaining health status parameters of multiple hybrid energy storage batteries at different charging states to form a charging information element matrix X1. The information elements of the row vector of the charging information element matrix X1 include: charging voltage U i1 , charging current A i1 , charging power W i1 , charging temperature T i1 , charging internal resistance R i1 , denoted as x i1 ={U i1 ,A i1 ,W i1 ,T i1 ,R i1}, the charging time x i1 Represents the health status data at different moments i;

[0010] Discharge sample information collection step: obtain the health status parameters of multiple hybrid energy storage batteries at different discharge times to form a discharge information element matrix X2. The information elements of the row vector of the discharge information element matrix X2 include: discharge voltage U i2 , discharge current A i2 , discharge power W i2 , discharge temperature T i2 , discharge internal resistance R i2 , denoted as x i2 ={U i2 ,A i2 ,W i2 ,T i2 ,R i2}, the x i2 Represents the health status data at different moments i during discharge;

[0011] Data sorting and fusion steps: Preprocess the health status data under charging state and the health status data under discharging state, including: S1, select the status data under the corresponding proportion space according to the preset threshold ratio, and delete the obviously unqualified data; S2, fill the empty value data by the average value method; S3, fuse the charging sample data and the discharging sample data at the corresponding time, and express it as Get the health status matrix X at the corresponding moment;

[0012] The step of estimating the SOH of the hybrid energy storage battery is as follows: inputting the data in the health state matrix X into a pre-trained health state model to predict the SOH of the hybrid energy storage battery;

[0013] The steps for judging the health level of the hybrid energy storage battery are as follows: according to the SOH of the hybrid energy storage battery, the corresponding health level is searched from the preset database.

[0014] Optionally, in a first implementation of the first aspect of the embodiment of the present invention, inputting the data in the health state matrix X into a pre-trained health state model to predict the SOH of the hybrid energy storage battery includes:

[0015] The hybrid verification method is used to predict the SOH of the current hybrid energy storage battery. The formula is: Among them, C now is the current capacity of the battery, C nominal is the rated capacity of the battery, R max ,R now , R are the maximum internal resistance, current internal resistance and initial internal resistance of the battery respectively, Q now ,Q nominal are the maximum capacity of the battery when it is fully charged and the rated capacity of the current battery, respectively. α, β, and χ represent the adjustment factors under the corresponding states.

[0016] Optionally, in a first implementation of the first aspect of the embodiment of the present invention, the health status model is obtained by training, specifically including:

[0017] Collect historical sample data uploaded by each edge node to the cloud server wirelessly; after data sorting and fusion processing of the historical sample data, calculate the aggregate value of the sample data of each dimension in the corresponding time window, and solve the average value of the aggregate value through the support vector;

[0018] Establish the corresponding relationship between the sample data of each dimension and store it in the MySQL database table; input the processed historical behavior data into the preset initial health status model;

[0019] The relevant adjustment factors are calculated through the initial health status model. On the basis of meeting the performance indicator threshold, the support vector machine method is used to train the health status model; the prediction results are continuously updated to the known performance indicator data sequence, and correlation analysis is performed. According to the degree of correlation, retraining is performed by expanding the training set to dynamically update the health status model.

[0020] Optionally, in a first implementation of the first aspect of the embodiment of the present invention, the aggregate value is averaged by solving the support vector, and the formula is:

[0021] average value where x i ,y i is a training sample, (xs ,y s ) is any support vector, S={i|a i >0,i=1,2,...,m} is the subscript set of all support vectors, a i is the Lagrange multiplier.

[0022] Optionally, in a first implementation of the first aspect of the embodiment of the present invention, the kernel function of the health status model trained by the support vector machine method is: Among them, K(x,x i )=Φ(x i ) T φ(x j ) is the kernel function, Φ(x i ),Φ(x j ) represent the samples x i ,x j The feature vector after mapping to the high-dimensional feature space, a i is the Lagrange multiplier, for a i Variable data, y i is the support vector training sample, w is the model parameter, and b is the mean value.

[0023] Optionally, in a first implementation of the first aspect of the embodiment of the present invention, searching a preset database for a corresponding health level according to the SOH of the hybrid energy storage battery includes:

[0024] A mapping relationship between the SOH data range and the health level is established in advance, the corresponding value range is found according to the SOH value obtained in the budget, and the health level assessment is obtained according to the data range through the mapping relationship.

[0025] A second aspect of an embodiment of the present invention provides a hybrid energy storage battery status monitoring system based on big data processing. The system is applied to the hybrid energy storage battery status monitoring method based on big data processing, and includes a data acquisition platform, wherein the data acquisition platform mainly includes: a hybrid energy storage battery pack, a wireless gateway, a cloud server, and an edge node; the system also includes:

[0026] Charging sample information collection module: obtains the health status parameters of multiple hybrid energy storage batteries at different charging states at different times, and forms a charging information element matrix X1. The information elements of the row vector of the information element matrix include: charging voltage U i1 , charging current A i1 , charging power W i1 , charging temperature T i1 , charging internal resistance R i1 , denoted as x i1 ={U i1,A i1 ,W i1 ,T i1 ,R i1}, the x i1 Represents the health status data at different moments i;

[0027] Discharge sample information collection module: obtains the health status parameters of multiple hybrid energy storage batteries at different discharge times to form a discharge information element matrix X2. The information elements of the row vector of the information element matrix include: discharge voltage U i2 , discharge current A i2 , discharge power W i2 , discharge temperature T i2 , discharge internal resistance R i2 , denoted as x i2 ={U i2 ,A i2 ,W i2 ,T i2 ,R i2}, the x i2 Represents the health status data at different moments i;

[0028] Data sorting and fusion module: pre-processing the health status data in the charging state and the health status data in the discharging state, including: S1, selecting the status data in the corresponding proportion space according to the preset threshold ratio, and deleting the obviously unqualified data; S2, filling the empty value data by the average value method; S3, fusing the charging sample data and the discharging sample data at the corresponding time, expressed as Get the health status matrix X at the corresponding moment;

[0029] Hybrid energy storage battery SOH estimation module: input the data in the health state matrix X into the pre-trained health state module to predict the SOH of the hybrid energy storage battery;

[0030] Health level evaluation module for hybrid energy storage batteries: According to the SOH of the hybrid energy storage battery, the corresponding health level is searched from the preset database.

[0031] A third aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored on the memory and runnable on the processor, wherein the processor implements the method for hybrid energy storage battery status monitoring based on big data processing when executing the computer program.

[0032] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium comprising instructions, which, when executed on a computer, enable the computer to execute any one of the methods for hybrid energy storage battery status monitoring based on big data processing.

[0033] The beneficial results of the above technical solution of this invention are as follows:

[0034] The technical solution provided in the embodiment of the present invention addresses the nonlinearity and difficulty in online evaluation of the hybrid energy storage battery state. A system is provided to sequentially collect charging sample information and discharging sample information, organize and fuse the data, estimate the state of health of the hybrid energy storage battery, and assess the health level of the hybrid energy storage battery to achieve business hybrid energy storage battery state monitoring. The sample data information elements include voltage, current, power, temperature, and internal resistance under charge and discharge conditions. By organizing and fusing the data and performing data prediction using a preset model, the hybrid energy storage battery state can be comprehensively covered, a comprehensive evaluation can be performed, and the accuracy of the evaluation can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings, which constitute a part of the present disclosure, are used to provide a further understanding of the present disclosure. The exemplary embodiments of the present disclosure and their descriptions are used to explain the present application and do not constitute an improper limitation to the present disclosure.

[0036] Figure 1 is a diagram of the overall hardware structure;

[0037] Figure 2 is a schematic diagram of the task scheduling process;

[0038] FIG3 is a schematic diagram of a task scheduling module.

[0039] Specific implementation methods

[0040] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0041] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used in the present disclosure have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

[0042] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0043] Embodiments of the present invention provide a hybrid energy storage battery status monitoring method and system based on big data processing, which are applied to the communication network architecture shown in Figure 1. The network structure includes a data acquisition platform, which primarily comprises a hybrid energy storage battery pack, a wireless gateway, a cloud server, and edge nodes. The edge nodes communicate with the cloud server platform via a wireless network. The cloud server is composed of a cluster of multiple edge nodes.

[0044] Optionally, the terminal node collects battery status parameters at different times through sensors and uploads them to the cloud server platform via a wireless gateway. The wireless gateway may include a Wireless Fidelity (WIFI) communication module, a Bluetooth (BLE) communication module, or a Zigbee communication module, and converts the data into wireless signals through the corresponding serial port for transmission.

[0045] Please refer to FIG2 , which is a flow chart of a hybrid energy storage battery status monitoring method based on big data processing provided by an embodiment of the present invention, which specifically includes:

[0046] Charging sample information collection step: obtain the health status parameters of multiple hybrid energy storage batteries at different charging states at different times to form a charging information element matrix X1. The information elements of the row vector of the information element matrix include: charging voltage U i1 , charging current A i1 , charging power W i1 , charging temperature T i1 , charging internal resistance R i1 , denoted as x i1 ={U i1 ,A i1 ,W i1 ,T i1 ,R i1}, the x i1 Represents the health status data at different moments i;

[0047] Discharge sample information collection step: obtain the health status parameters of multiple hybrid energy storage batteries at different discharge times to form a discharge information element matrix X2. The information elements of the row vector of the information element matrix include: discharge voltage U i2 , discharge current A i2 , discharge power W i1 , discharge temperature T i2 , discharge internal resistance R i2 , denoted as x i2 ={U i2 ,A i2 ,W i2 ,T i2 ,R i2}, the x i2 Represents the health status data at different moments i;

[0048] Data sorting and fusion step: pre-processing the health status data in the charging state and the health status data in the discharging state, including: S1, selecting the status data in the corresponding proportion space according to the preset threshold ratio, and deleting the obviously unqualified data; S2, filling the null value data by the average value method; S3, fusing the charging sample data and the discharging sample data at the corresponding time, expressed as Get the health status matrix X at the corresponding moment;

[0049] The step of estimating the SOH of the hybrid energy storage battery is as follows: inputting the data in the health state matrix X into a pre-trained health state model to predict the SOH of the hybrid energy storage battery;

[0050] The steps for judging the health level of the hybrid energy storage battery are as follows: according to the SOH of the hybrid energy storage battery, the corresponding health level is searched from the preset database.

[0051] S110, charging sample information collection step: obtaining health status parameters of multiple hybrid energy storage batteries at different charging states at different times, forming a charging information element matrix X1, the information elements of the row vector of the information element matrix include: charging voltage U i1 , charging current A i1 , charging power W i1 , charging temperature T i1 , charging internal resistance R i1 , denoted as x i1 ={U i1 ,A i1 ,W i1 ,T i1 ,R i1}, the x i1 Represents the health status data at different times i.

[0052] Optionally, although the charging state has stable voltage and current, due to the difference in ambient temperature, battery temperature, and the different degrees of attenuation of the battery itself, the measured data does not have a stable linear relationship. Therefore, it is necessary to measure the charging voltage U i1 , charging current A i1 , charging power W i1 , charging temperature T i1 , charging internal resistance R i1 Collecting data helps optimize the prediction model.

[0053] S112, discharge sample information collection step: obtain health status parameters of multiple hybrid energy storage batteries at different discharge times to form a discharge information element matrix X2, the information elements of the row vector of the information element matrix include: discharge voltage U i2 , discharge current A i2 , discharge power W i1 , discharge temperature T i2 , discharge internal resistance R i2 , denoted as x i2 ={U i2 ,A i2 ,W i2 ,T i2 ,R i2}, the x i2 Represents the health status data at different times i.

[0054] Optionally, to study the discharge state of the battery at different currents, constant current discharge is performed using currents of 5 A, 10 A, and 15 A. The voltage, current, temperature, and other information during the discharge process are sampled at a frequency of 30 seconds.

[0055] S113, data collating and fusion step: pre-processing the health status data in the charging state and the health status data in the discharging state, including: S3.1, selecting the status data in the corresponding proportion space according to the preset threshold ratio, and deleting the obviously unqualified data; S3.2, filling the null value data by the average value method; S3, fusing the charging sample data and the discharging sample data at the corresponding time, expressed as Get the health status matrix X at the corresponding moment.

[0056] S114, a step of estimating the SOH of the hybrid energy storage battery: inputting the data in the health state matrix X into a pre-trained health state model to predict the SOH of the hybrid energy storage battery.

[0057] Optionally, the data in the health state matrix X is input into a pre-trained health state model to predict the SOH of the hybrid energy storage battery, including: using a hybrid verification method to predict the SOH of the current hybrid energy storage battery, the formula is:

[0058] Among them, C now is the current capacity of the battery, C nominal is the rated capacity of the battery, R max ,R now , R are the maximum internal resistance, current internal resistance and initial internal resistance of the battery respectively, Q now ,Q nominalare the maximum capacity of the battery when it is fully charged and the rated capacity of the current battery, respectively. α, β, and χ represent the adjustment factors under the corresponding states.

[0059] Preferably, the health status model is obtained through training, specifically including: S4.1 wirelessly collecting historical sample data uploaded by each edge node to the cloud server; S4.2 performing data sorting and fusion processing on the historical sample data, calculating the aggregation value of the sample data of each dimension in the corresponding time window, and solving the average value of the aggregation value through the support vector; S4.3 establishing the correspondence between the sample data of each dimension, and storing it in the MySQL database form; S4.4 inputting the processed historical behavior data into the preset initial health status model; S4.5 calculating the relevant adjustment factors through the initial health status model, and using the support vector machine method to train the health status model on the basis of meeting the performance index threshold; S4.6 continuously updating the prediction results to the known performance index data sequence, and performing correlation analysis. According to the degree of correlation, retraining is performed by expanding the training set to dynamically update the health status model.

[0060] Preferably, the initial health status model is pre-established based on a bidirectional long short-term memory (LSTM) recurrent neural network.

[0061] Optionally, the aggregated values ​​are averaged by the support vectors using the formula: where x i ,y i is a training sample, (x s ,y s ) is any support vector, S={i|a i >0,i=1,2,...,m} is the subscript set of all support vectors, a i is the Lagrange multiplier.

[0062] Optionally, the support vector machine method is used to train the health status model. The kernel function is: Among them, K(x,x i )=Φ(x i ) T φ(x j ) is the kernel function, Φ(x i ),Φ(x j ) represent the samples x i ,x j The feature vector after mapping to the high-dimensional feature space, a i is the Lagrange multiplier, for a i Variable data, x i ,y iis the support vector training sample, w is the model parameter, and b is the mean value.

[0063] S115, step of judging the health level of the hybrid energy storage battery: searching for the corresponding health level from a preset database according to the SOH of the hybrid energy storage battery.

[0064] Optionally, the method of searching for a corresponding health level from a preset database based on the SOH of the hybrid energy storage battery includes: pre-establishing a mapping relationship between the SOH data range and the health level, searching for a corresponding numerical range based on the SOH value obtained in the budget, and obtaining a health level assessment based on the data range through the mapping relationship.

[0065] An embodiment of the present invention further provides a hybrid energy storage battery status monitoring system based on big data processing, which is applied to a cloud-coordinated communication network based on platform service types as shown in FIG1 .

[0066] As shown in FIG3 , the system includes: a charging sample information collection module 1 , a discharging sample information collection module 2 , a data sorting and fusion module 3 , a hybrid energy storage battery SOH estimation module 4 , and a hybrid energy storage battery health level assessment step 5 .

[0067] The charging sample information collection module 1 is used to obtain the health status parameters of multiple hybrid energy storage batteries at different charging states at different times, and form a charging information element matrix X1. The information elements of the row vector of the information element matrix include: charging voltage U i1 , charging current A i1 , charging power W i1 , charging temperature T i1 , charging internal resistance R i1 ;

[0068] The discharge sample information collection module 2 is used to obtain the health status parameters of multiple hybrid energy storage batteries at different discharge times to form a discharge information element matrix X2. The information elements of the row vector of the information element matrix include: discharge voltage U i2 , discharge current A i2 , discharge power W i1 , discharge temperature T i2 , discharge internal resistance R i2 ;

[0069] The data sorting and fusion module 3 is used to pre-process the health status data in the charging state and the health status data in the discharging state, including: S1, selecting the status data in the corresponding proportion space according to the preset threshold ratio, and deleting the data that is obviously unqualified; S2, filling the empty value data by the average value method; S3, fusing the charging sample data and the discharging sample data at the corresponding time;

[0070] The hybrid energy storage battery SOH estimation module 4 is used to input the data in the health state matrix X into a pre-trained health state module to predict the SOH of the hybrid energy storage battery; the hybrid energy storage battery health level evaluation step 5 is used to search the corresponding health level from a preset database based on the SOH of the hybrid energy storage battery.

[0071] An embodiment of the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for monitoring the status of a hybrid energy storage battery based on big data processing is implemented.

[0072] An embodiment of the present invention also provides a computer-readable storage medium comprising instructions, which, when executed on a computer, enable the computer to execute the method for hybrid energy storage battery status monitoring based on big data processing.

[0073] Although the above describes the specific implementation methods of the present disclosure in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present disclosure. Those skilled in the art should understand that on the basis of the technical solution of the present disclosure, various modifications or variations that can be made by those skilled in the art without creative work are still within the scope of protection of the present disclosure.

Claims

1. A hybrid energy storage battery status monitoring method based on big data processing, characterized in that: A hybrid energy storage battery state monitoring system applied to big data processing, the method comprising: a charging sample information collection step: obtaining health state parameters of multiple hybrid energy storage batteries under different charging states at different times, forming a charging information element matrix X1, wherein the information elements of the row vector of the charging information element matrix X1 include: charging voltage U i1 , Charging current A i1 , Charging power W i1 , Charging temperature T i1 、Charging internal resistance R i1 , denoted by x i1 = {U i1 ,A i1 ,W i1 ,T i1 ,R i1 }, x i1 Represents the health status data at different times i during charging; Discharge sample information collection step: obtain the health status parameters of multiple hybrid energy storage batteries at different discharge states at different times to form a discharge information element matrix X2. The information elements of the row vector of the discharge information element matrix X2 include: discharge voltage U i2 , discharge current A i2 , discharge power W i2 , discharge temperature T i2 , discharge internal resistance R i2 , denoted by x i2 = {U i2 ,A i2 ,W i2 ,T i2 ,R i2 }, x i2 Represents the health status data at different times i during discharge; Data sorting and fusion steps: pre-process the health status data under charging state and the health status data under discharging state, including: S1, select the state data under the corresponding proportion space according to the preset threshold ratio, and delete the obviously unqualified data; S2, fill the empty value data by the average value method; S3, fuse the charging sample data and the discharging sample data at the corresponding time, expressed as Get the health status matrix X at the corresponding time; The steps for estimating the SOH of the hybrid energy storage battery are as follows: the data in the health state matrix X is input into the pre-trained health state model to predict the SOH of the hybrid energy storage battery; The steps for judging the health level of the hybrid energy storage battery are as follows: according to the SOH of the hybrid energy storage battery, the corresponding health level is searched from the preset database.

2. A hybrid energy storage battery status monitoring method based on big data processing according to claim 1, characterized in that: The data in the health state matrix X is input into a pre-trained health state model to predict the SOH of the hybrid energy storage battery, including: The hybrid verification method is used to predict the SOH of the current hybrid energy storage battery. The formula is: Among them, C now is the current capacity of the battery, C nominal is the rated capacity of the battery, R max ,R now , R are the maximum internal resistance, current internal resistance and initial internal resistance of the battery respectively, Q now ,Q nominal They are the maximum capacity of the battery when it is fully charged and the rated capacity of the current battery, respectively. α, β, and χ represent the adjustment factors under the corresponding states.

3. The hybrid energy storage battery status monitoring method based on big data processing according to claim 1 is characterized in that: The health status model is obtained by training, specifically including: Collect historical sample data uploaded by each edge node to the cloud server wirelessly; After data collating and fusion processing of historical sample data, the aggregate value of each dimension sample data in the corresponding time window is calculated, and the aggregate value is obtained by solving the average value of the support vector; the corresponding relationship between the sample data of each dimension is established and stored in the MySQL database form; the processed historical behavior data is input into the preset initial health status model; The relevant adjustment factors are calculated through the initial health status model. On the basis of meeting the performance indicator threshold, the support vector machine method is used to train the health status model. The prediction results are continuously updated to the known performance indicator data sequence, and correlation analysis is performed. According to the degree of correlation, retraining is performed by expanding the training set to dynamically update the health status model.

4. A hybrid energy storage battery status monitoring method based on big data processing according to claim 3, characterized in that: The aggregate value is obtained by solving the average value of the support vector, and the formula is: average value where x i ,y i is a training sample, (x s ,y s ) is an arbitrary support vector, S = {i|a i >0,i=1,2,...,m} is the subscript set of all support vectors, a i is the Lagrange multiplier.

5. A hybrid energy storage battery status monitoring method based on big data processing according to claim 4, characterized in that: The kernel function of the support vector machine method for training the health status model is: K(x,x i )=Φ(x i ) T φ(x j ) is the kernel function, Φ(x i ),Φ(x j ) represent the samples x i ,x j The feature vector after mapping to the high-dimensional feature space, a i is the Lagrange multiplier, for a i The variable data, y i is the support vector training sample, w is the model parameter, and b is the mean value.

6. The hybrid energy storage battery status monitoring method based on big data processing according to claim 1 is characterized in that: According to the SOH of the hybrid energy storage battery, searching for the corresponding health level from a preset database includes: A mapping relationship between the SOH value range and the health level is established in advance, the corresponding value range is searched according to the pre-obtained SOH value, and the health level assessment is obtained according to the mapping relationship based on the value range.

7. A hybrid energy storage battery status monitoring system based on big data processing, the system is applied to the hybrid energy storage battery status monitoring method based on big data processing according to claim 1, comprising a data acquisition platform, wherein: The data collection platform includes: hybrid energy storage battery pack, wireless gateway, cloud server, edge node; the system also includes: Charging sample information collection module: obtain the health status parameters of multiple hybrid energy storage batteries at different charging states at different times to form a charging information element matrix X1. The information elements of the row vector of the charging information element matrix X1 include: charging voltage U i1 , Charging current A i1 , Charging power W i1 , Charging temperature T i1 、Charging internal resistance R i1 , denoted by x i1 = {U i1 ,A i1 ,W i1 ,T i1 ,R i1 }, the x i1 Represents the health status data at different times i during charging; Discharge sample information collection module: obtain the health status parameters of multiple hybrid energy storage batteries at different times of discharge, and form a discharge information element matrix X2. The information elements of the row vector of the discharge information element matrix include: discharge voltage U i2 , discharge current A i2 , discharge power W i2 , discharge temperature T i2 , discharge internal resistance R i2 , denoted by x i2 = {U i2 ,A i2 ,W i2 ,T i2 ,R i2 }, the x i2 Represents the health status data at different times i during discharge; Data sorting and fusion module: pre-process the health status data under charging state and the health status data under discharging state, including: S1, select the state data under the corresponding proportion space according to the preset threshold ratio, and delete the obviously unqualified data; S2, fill the empty value data by the average value method; S3, fuse the charging sample data and the discharging sample data at the corresponding time, expressed as Get the health status matrix X at the corresponding time; Hybrid energy storage battery SOH estimation module: input the data in the health state matrix X into a pre-trained health state module to predict the SOH of the hybrid energy storage battery; Hybrid energy storage battery health level evaluation module: According to the SOH of the hybrid energy storage battery, The corresponding health level is searched in the given database.

8. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements a hybrid energy storage battery status monitoring method based on big data processing as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: It includes instructions, which, when executed on a computer, enable the computer to execute a hybrid energy storage battery status monitoring method based on big data processing as described in any one of claims 1 to 6.

Images