Method for predicting service life of energy storage battery in industrial and commercial energy storage system

Abstract

The invention provides a method for predicting the service life of an energy storage battery in an industrial and commercial energy storage system, and the method comprises the steps: determining a battery aging sensitive scale in each scale space according to the sample entropy and permutation entropy of a voltage component of a to-be-detected energy storage battery in a multi-scale space, and carrying out the health assessment of the battery aging sensitive scale. Obtaining a scale entropy health index of the to-be-tested energy storage battery; extracting the maximum surface temperature of the to-be-measured energy storage battery at the end of the constant current stage and the equilibrium temperature after standing for a specified time period from the battery temperature data, and determining the thermal relaxation equilibrium index of the to-be-measured energy storage battery in the charging process based on the maximum surface temperature and the equilibrium temperature; and performing combined prediction on the cycle life of the to-be-tested energy storage battery through the scale entropy health index and the thermal relaxation balance index to obtain a combined prediction value of the residual cycle life of the to-be-tested energy storage battery. Based on the scheme, multi-dimensional collaborative evaluation of the electrochemical state and the thermal management performance of the energy storage battery in the industrial and commercial energy storage system can be realized.

Description

Technical Field

[0001] This application relates to the technical field of energy storage batteries. More specifically, this application relates to a method for predicting the life of energy storage batteries in industrial and commercial energy storage systems. Background Art

[0002] In the industrial and commercial fields, the application demand for energy storage batteries has increased significantly, and significant progress has been made in energy storage battery technology. For example, lithium-ion batteries are developing towards large capacity and long life, and sodium-ion batteries are gradually achieving industrialization. When the existing technology uses a single index such as capacity attenuation or internal resistance growth to predict the battery life, the essence of capacity attenuation is the loss of the total amount of lithium-ion active substances and recyclable lithium, but it cannot distinguish the contribution differences between the phase change of the positive electrode material (electrochemical aging) and the drying of the electrolyte (thermal management failure). Although the internal resistance growth can reflect the interfacial impedance, it confuses the influence mechanisms of film thickening (electrochemical side reactions) and tab loosening (mechanical connection problems). This coupling effect causes the traditional method to accelerate capacity attenuation due to thermal management failure in a high-temperature environment, but it cannot quantify the separate influence of the decrease in the efficiency of the cooling system. During low-temperature charging, lithium deposition and increased electrolyte viscosity jointly push up the internal resistance. Therefore, how to achieve multi-dimensional collaborative evaluation of the electrochemical state and thermal management performance of energy storage batteries in industrial and commercial energy storage systems, and then improve the accuracy of energy storage battery life prediction has become a difficult problem faced by the industry. Summary of the Invention

[0003] This application provides a method for predicting the life of energy storage batteries in industrial and commercial energy storage systems, which can achieve multi-dimensional collaborative evaluation of the electrochemical state and thermal management performance of energy storage batteries in industrial and commercial energy storage systems, and then improve the accuracy of energy storage battery life prediction.

[0004] In a first aspect, this application provides a method for predicting the life of energy storage batteries in industrial and commercial energy storage systems, including: Connect the energy storage battery to be tested in the industrial and commercial energy storage system to a standard charging device, and collect the voltage time series signal and battery temperature data of the energy storage battery to be tested during multiple charging processes; Perform linear interpolation on the voltage time series signal, and then extract the sample entropy and permutation entropy of the voltage components in the multi-scale space. Determine the battery aging sensitive scale in each scale space according to the sample entropy and permutation entropy of each voltage component, and perform a health assessment on the battery aging sensitive scale based on the multi-scale voltage trajectory matrix of the energy storage battery to be tested during the charging process to obtain the scale entropy health index of the energy storage battery to be tested; Extract the surface maximum temperature of the energy storage battery to be tested at the end of the constant current stage and the equilibrium temperature after standing for a specified period of time from the battery temperature data, and then determine the thermal relaxation equilibrium index of the energy storage battery to be tested during the charging process based on the surface maximum temperature and the equilibrium temperature; The cycle life of the energy storage battery to be measured is jointly predicted by the scale entropy health index and the thermal relaxation balance index, and a joint prediction value of the remaining cycle life in the energy storage battery to be measured is obtained.

[0005] In some embodiments, linear interpolation is performed on the voltage time series signal, and then extracting the sample entropy and permutation entropy of the voltage components in the multi-scale space specifically includes: Perform multi-layer decomposition on the voltage time series signal to obtain multiple scale spaces, and then use cubic spline interpolation on the voltage time series signal to a specified sampling rate; Calculate the voltage components of the specified time window in each scale space through the voltage time series signal after spline interpolation; Determine the sample entropy and permutation entropy of the voltage components in the multi-scale space through all the voltage components.

[0006] In some embodiments, determining the battery aging sensitive scale in each scale space according to the sample entropy and permutation entropy of each voltage component specifically includes: For each scale space, obtain the initial sample entropy and initial permutation entropy of the energy storage battery to be measured in the scale space; Determine the sample entropy decay rate of the voltage components in the scale space through the initial sample entropy and the sample entropy, and at the same time determine the permutation entropy decay rate of the voltage components in the scale space through the initial permutation entropy and the permutation entropy; Determine the battery aging sensitive scale in the scale space according to the sample entropy decay rate and the permutation entropy decay rate, and then obtain the battery aging sensitive scale in each scale space.

[0007] In some embodiments, based on the multi-scale voltage trajectory matrix during the charging process of the energy storage battery to be measured, health assessment is performed on the battery aging sensitive scale, and obtaining the scale entropy health index of the energy storage battery to be measured specifically includes: Construct a voltage trajectory matrix including sensitive scale components based on the voltage time series signal; Extract the contribution rate of the first principal component variance in the voltage trajectory matrix through principal component analysis as the health benchmark value; Determine the scale entropy health index of the energy storage battery to be measured according to the health benchmark value and the battery aging sensitive scale.

[0008] In some embodiments, extracting the maximum surface temperature at the end of the constant current stage and the equilibrium temperature after standing for a specified period of time from the battery temperature data of the energy storage battery to be measured specifically includes: Obtain multiple surface temperatures of the energy storage battery to be measured at the end of the constant current stage and multiple standing temperatures after standing for a specified period of time from the battery temperature data; Select the maximum surface temperature of the energy storage battery to be measured at the end of the constant current stage from all the surface temperatures; Determine the equilibrium temperature of the energy storage battery to be measured after a specified period of standing through all the standing temperatures.

[0009] In some embodiments, determining the thermal relaxation equilibrium index of the energy storage battery to be measured during charging based on the maximum surface temperature and the equilibrium temperature specifically includes: Obtain the thermal resistance correction coefficient and the standing time of the energy storage battery to be measured; Determine the temperature decay rate of the energy storage battery to be measured during charging through the maximum surface temperature, the standing time, and the equilibrium temperature; Determine the thermal relaxation equilibrium index of the energy storage battery to be measured during charging according to the temperature decay rate and the thermal resistance correction coefficient.

[0010] In some embodiments, jointly predicting the cycle life of the energy storage battery to be measured through the scale entropy health index and the thermal relaxation equilibrium index to obtain the joint prediction value of the remaining cycle life in the energy storage battery to be measured specifically includes: Initialize an ensemble prediction model based on boosting trees; Use the scale entropy health index as the voltage feature in the ensemble prediction model; Use the thermal relaxation equilibrium index as the heat dissipation feature in the ensemble prediction model; Use the ensemble prediction model to predict the cycle life of the energy storage battery to be measured to obtain the joint prediction value of the remaining cycle life in the energy storage battery to be measured.

[0011] In some embodiments, use a high-precision data acquisition card to collect the voltage time series signal of the energy storage battery to be measured during multiple charging processes.

[0012] In a second aspect, the present application provides a computer device, which includes a memory and a processor. The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-mentioned method for predicting the life of the energy storage battery in the industrial and commercial energy storage system.

[0013] In a third aspect, the present application provides a computer-readable storage medium, in which instructions or codes are stored. When the instructions or codes are run on a computer, the computer is caused to execute the above-mentioned method for predicting the life of the energy storage battery in the industrial and commercial energy storage system.

[0014] The technical solutions provided by the disclosed embodiments of the present application have the following beneficial effects: In a method for predicting the life of energy storage batteries in an industrial and commercial energy storage system provided by this application, the energy storage battery to be tested in the industrial and commercial energy storage system is connected to a standard charging device, and the voltage time series signal and battery temperature data during multiple charging processes of the energy storage battery to be tested are collected; linear interpolation is performed on the voltage time series signal, and then the sample entropy and permutation entropy of the voltage components in the multi-scale space are extracted. According to the sample entropy and permutation entropy of each voltage component, the battery aging sensitive scale in each scale space is determined. Based on the multi-scale voltage trajectory matrix of the energy storage battery to be tested during the charging process, health assessment is performed on the battery aging sensitive scale to obtain the scale entropy health index of the energy storage battery to be tested; the surface maximum temperature at the end of the constant current stage and the equilibrium temperature after standing for a specified period of time are extracted from the battery temperature data of the energy storage battery to be tested, and then the thermal relaxation equilibrium index of the energy storage battery to be tested during the charging process is determined based on the surface maximum temperature and the equilibrium temperature; the cycle life of the energy storage battery to be tested is jointly predicted through the scale entropy health index and the thermal relaxation equilibrium index, and the joint prediction value of the remaining cycle life in the energy storage battery to be tested is obtained.

[0015] It can be seen that in this application, the cycle life of the energy storage battery to be tested is jointly predicted by the scale entropy health index and the thermal relaxation balance index, and the joint prediction value of the remaining cycle life in the energy storage battery to be tested is obtained. First, determining the scale entropy health index can obtain the quantitative evaluation parameters of the battery's electrochemical state, thereby realizing the accurate diagnosis of the internal aging mechanism of the energy storage battery. Through multi-scale voltage signal decomposition and entropy value analysis, the subtle changes in the voltage waveforms of different frequency bands can be captured. The increase in sample entropy at high frequencies reflects the aggravation of electrochemical noise caused by the precipitation of lithium dendrites, and the decrease in permutation entropy at low frequencies indicates the distortion of reaction kinetics caused by the phase change of the electrode material, enabling the initial aging characteristics to be detected in advance through cyclic detection. By establishing a voltage trajectory matrix and principal component analysis, the multi-dimensional entropy change information is further fused into a single health score, enabling maintenance personnel to intuitively judge whether the battery is in uniform aging or local deterioration, thereby improving the life prediction accuracy of the energy storage battery. Then, determining the thermal relaxation balance index can obtain the dynamic evaluation index of the battery thermal management system's effectiveness, thereby revealing the influence mechanism of temperature factors on battery aging. By analyzing the temperature decay characteristics after the end of constant current charging, the ability of the battery to recover from the working state to the thermal equilibrium state is quantified. The surface maximum temperature reflects the instantaneous thermal load intensity, and the equilibrium temperature reflects the system's heat dissipation ability. The difference between the two combined with the thermal resistance correction coefficient can accurately evaluate the performance degradation of the cooling system. This thermal relaxation balance index is suitable for identifying hidden thermal risks that are difficult to detect by traditional methods. Through spatio-temporal correlation analysis with the scale entropy health index, it can also distinguish whether the capacity decay is due to the degradation of the bulk material or the failure of the thermal management, thereby further improving the life prediction accuracy of the energy storage battery. In summary, based on the above solutions, multi-dimensional collaborative evaluation of the electrochemical state and thermal management performance of energy storage batteries in industrial and commercial energy storage systems can be realized, thereby improving the life prediction accuracy of energy storage batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 is an exemplary flowchart of a method for predicting the life of an energy storage battery in an industrial and commercial energy storage system shown in some embodiments of the present application; Figure 2 is a schematic flowchart of realizing joint prediction shown in some embodiments of the present application; Figure 3 is a schematic structural diagram of a computer device for realizing a method for predicting the life of an energy storage battery in an industrial and commercial energy storage system shown in some embodiments of the present application. Detailed implementation manners

[0018] To better understand the technical solution of the present application, the technical solution of the present application will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0019] Refer to Figure 1 , which is an exemplary flowchart of a method for predicting the life of energy storage batteries in an industrial and commercial energy storage system according to some embodiments of the present application. The method for predicting the life of energy storage batteries in the industrial and commercial energy storage system mainly includes the following steps: In step 101, the energy storage battery to be tested in the industrial and commercial energy storage system is connected to a standard charging device, and the voltage time series signal and battery temperature data of the energy storage battery to be tested during multiple charging processes are collected.

[0020] It should be noted that in the present application, the battery temperature data is used to monitor the thermal behavior evolution during the charge and discharge process; the voltage time series signal is a high-resolution time series reflecting the internal electrochemical dynamics of the battery; the energy storage battery to be tested is a ternary lithium iron phosphate battery pack; the standard charging device is a programmable DC power supply supporting mode switching; specifically, when implemented, the energy storage battery to be tested is connected to an industrial-grade charging device with a constant current-constant voltage (CC-CV) mode to ensure that the charging parameters meet the battery technical specifications. The voltage time series signal (sampling rate ≥ 1Hz) and distributed temperature sensor (such as: surface-mounted thermocouple or infrared temperature measurement) data of the entire charging process are synchronously recorded through a high-precision data acquisition system (for example: 24-bit ADC module), covering the key temperature measurement points (positive and negative electrodes, middle of the housing, etc.) on the battery surface, and the voltage time series signal and battery temperature data of the energy storage battery to be tested during multiple charging processes can be obtained.

[0021] In step 102, linear interpolation is performed on the voltage time series signal, and then the sample entropy and permutation entropy of the voltage components in the multi-scale space are extracted. According to the sample entropy and permutation entropy of each voltage component, the battery aging sensitive scale in each scale space is determined, and the battery aging sensitive scale is health-assessed based on the multi-scale voltage trajectory matrix of the energy storage battery to be tested during the charging process, and the scale entropy health index of the energy storage battery to be tested is obtained.

[0022] In some embodiments, the linear interpolation of the voltage time series signal and the extraction of the sample entropy and permutation entropy of the voltage components in the multi-scale space can be implemented by the following steps: Perform multi-layer decomposition on the voltage time series signal to obtain multiple scale spaces, and then perform cubic spline interpolation on the voltage time series signal to a specified sampling rate; Calculate the voltage components of a specified time window in each scale space through the voltage time series signal after spline interpolation; Determine the sample entropy and permutation entropy of the voltage components in the multi-scale space through all the voltage components.

[0023] It should be noted that in this application, sample entropy is a parameter quantifying the randomness of voltage waveforms; permutation entropy is a parameter quantifying the orderliness of voltage waveforms; the specified sampling rate is defaulted to 1 Hz; the voltage component characterizes the local statistical characteristics of the voltage waveform in a specific frequency band.

[0024] In specific implementation, first, discrete wavelet transform (e.g., Db4 wavelet basis) is used to decompose the voltage time series signal into 4 layers to obtain a multi-scale space containing high-frequency detail components (D1 - D4) and low-frequency approximation components (A4). For voltage time series signals with insufficient sampling rate, a continuous smooth curve is constructed in the time domain through cubic spline interpolation and resampled at 1 Hz intervals to ensure the integrity of high-frequency components. Then, for each scale space, a 100-point time window slides in the scale space (e.g., D3 corresponds to the 2 - 4 Hz frequency band), and the voltage sequence within the time window is intercepted. After eliminating the boundary effect through zero-phase filtering, the voltage mean, standard deviation, and differential extreme value within the window are calculated as the voltage components of the specified time window (defaulted to 100 points) in the scale space. In this way, the voltage components of the specified time window in each scale space can be obtained. Finally, for the voltage components in each scale space, 0.2 times the standard deviation of the voltage component is used as the sample entropy of the voltage component in the scale space, and the entropy value with an embedding dimension of 3 and a delay of 1 in the voltage component is used as the permutation entropy of the voltage component in the scale space. In this way, the sample entropy and permutation entropy of the voltage component in the multi-scale space can be obtained.

[0025] In some embodiments, determining the battery aging sensitive scale in each scale space according to the sample entropy and permutation entropy of each voltage component can be implemented by the following steps: For each scale space, obtain the initial sample entropy and initial permutation entropy of the energy storage battery to be tested in the scale space; Determine the sample entropy decay rate of the voltage component in the scale space through the initial sample entropy and the sample entropy, and at the same time determine the permutation entropy decay rate of the voltage component in the scale space through the initial permutation entropy and the permutation entropy; Determine the battery aging sensitive scale in the scale space according to the sample entropy decay rate and the permutation entropy decay rate, and then obtain the battery aging sensitive scale in each scale space.

[0026] It should be noted that in this application, the battery aging sensitive scale; the initial sample entropy is the sample entropy of the battery in a brand-new state, and the initial permutation entropy is the permutation entropy of the battery in a brand-new state; the sample entropy decay rate represents the deterioration degree of the battery's electrochemical noise; the permutation entropy decay rate represents the degradation degree of the regularity of the voltage curve.

[0027] In specific implementation, first, for each scale space, obtain the initial sample entropy and initial permutation entropy of the energy storage battery under test in the scale space from the user manual of the energy storage battery under test; then, use the ratio of the result of subtracting the sample entropy from the initial sample entropy to the initial sample entropy as the sample entropy decay rate of the voltage component in the scale space, and use the ratio of the result of subtracting the permutation entropy from the initial permutation entropy to the initial sample entropy as the permutation entropy decay rate of the voltage component in the scale space; finally, set the decay threshold by combining the user manual of the energy storage battery under test with historical experience. The sample entropy decay threshold is defaulted to 15%, and the permutation entropy decay threshold is defaulted to 10%. If the sample entropy decay rate and permutation entropy decay rate of the scale space both satisfy being greater than the decay threshold, it is determined as a sensitive scale (for example: the high-frequency scale is usually sensitive to lithium deposition, and the low-frequency scale is sensitive to capacity loss). Take the output sensitive scale list and its decay weight as the battery aging sensitive scale under the scale space. Through the above method, the battery aging sensitive scale under each scale space can be obtained.

[0028] In some embodiments, based on the multi-scale voltage trajectory matrix during the charging process of the energy storage battery under test, the health assessment of the battery aging sensitive scale can be realized by the following steps to obtain the scale entropy health index of the energy storage battery under test: Construct a voltage trajectory matrix containing sensitive scale components based on the voltage time series signal; Extract the contribution rate of the first principal component variance in the voltage trajectory matrix through principal component analysis as the health benchmark value; Determine the scale entropy health index of the energy storage battery under test according to the health benchmark value and the battery aging sensitive scale.

[0029] It should be noted that in this application, the scale entropy health index is a composite health score for the battery life prediction model; the voltage trajectory matrix is a structured data set reflecting the evolution of multi-scale voltage characteristics during battery aging; the health benchmark value is the core index characterizing the comprehensive aging state of the battery.

[0030] In specific implementation, first, based on the identified battery aging sensitive scales (e.g., high-frequency scale D3, low-frequency scale A4), characteristic parameters of each scale component are obtained from the voltage time-series signal. The characteristic parameters include time-domain statistics (e.g., mean, variance), frequency-domain energy ratio, and entropy decay rate. All the characteristic parameters are classified by scale and organized into a multi-dimensional matrix. The rows of the matrix represent different charge cycles, and the columns represent the characteristic values of each scale component, thus obtaining a voltage trajectory matrix containing sensitive scale components. Then, based on the principal component analysis algorithm, the variance contribution rate of each principal component in the voltage trajectory matrix is calculated. Since the first principal component usually contains the most significant degradation trend information in the data (e.g., capacity decay, internal resistance increase), the variance contribution rate of the first principal component is selected as the health benchmark value. Finally, the health benchmark value is weighted and fused with the entropy decay values of each sensitive scale in the battery aging sensitive scales. The weight assignment is based on the physical meaning of the sensitive scales (e.g., the high-frequency scale is sensitive to lithium deposition and is assigned a higher weight), so as to use the result of the weighted fusion as the scale entropy health index. Among them, the value range of the scale entropy health index is within 0 - 1. 1 indicates healthy, and below 0.8 indicates significant aging.

[0031] In step 103, the surface maximum temperature of the energy storage battery to be tested at the end of the constant current stage and the equilibrium temperature after standing for a specified period of time are extracted from the battery temperature data, and then the thermal relaxation equilibrium index of the energy storage battery to be tested during the charging process is determined based on the surface maximum temperature and the equilibrium temperature.

[0032] In some embodiments, the extraction of the surface maximum temperature of the energy storage battery to be tested at the end of the constant current stage and the equilibrium temperature after standing for a specified period of time from the battery temperature data can be implemented by the following steps: Obtain multiple surface temperatures of the energy storage battery to be tested at the end of the constant current stage and multiple standing temperatures after standing for a specified period of time from the battery temperature data; Screen out the surface maximum temperature of the energy storage battery to be tested at the end of the constant current stage from all the surface temperatures; Determine the equilibrium temperature of the energy storage battery to be tested after standing for a specified period of time through all the standing temperatures.

[0033] It should be noted that in this application, the equilibrium temperature is the stable temperature of the battery without external thermal disturbances; the maximum surface temperature is the peak indicator of the battery's thermal behavior during charging. Specifically, in implementation, first, obtain multiple surface temperatures of the energy storage battery under test at the end of the constant current stage and multiple rest temperatures after a specified rest period from the battery temperature data. The surface temperature refers to the instantaneous temperature of the battery at the end of charging, and the rest temperature refers to the temperature after a specified rest period. Then, take the maximum value among all the surface temperatures as the maximum surface temperature of the energy storage battery under test at the end of the constant current stage. Finally, if the temperature rises by more than 1°C during the rest period, extend the rest to 20 minutes, and take the minimum value among all the rest temperatures as the equilibrium temperature of the energy storage battery under test after the specified rest period.

[0034] In some embodiments, the thermal relaxation equilibrium index of the energy storage battery under test during charging can be determined based on the maximum surface temperature and the equilibrium temperature by the following steps: Obtain the thermal resistance correction coefficient and the rest time of the energy storage battery under test; Determine the temperature decay rate of the energy storage battery under test during charging through the maximum surface temperature, the rest time, and the equilibrium temperature; Determine the thermal relaxation equilibrium index of the energy storage battery under test during charging according to the temperature decay rate and the thermal resistance correction coefficient.

[0035] It should be noted that in this application, the thermal relaxation equilibrium index is a dynamic index for comprehensively evaluating the battery's thermal recovery ability; the thermal resistance correction coefficient is used to calibrate the deviation between the actual heat dissipation ability of the battery and the design value; the rest time is the minimum observation duration required to ensure that the temperature reaches a stable state; the temperature decay rate is to quantify the speed at which the battery recovers from the working state to the thermal equilibrium state.

[0036] Specifically, in implementation, first, obtain the single rest time of the energy storage battery under test during multiple charging processes, and thus take the average value of all single rest times as the rest time of the energy storage battery under test. At the same time, the thermal resistance correction coefficient can be obtained from the user manual of the energy storage battery under test. Then, the ratio of the result of subtracting the equilibrium temperature from the maximum surface temperature to the rest time can be used as the temperature decay rate of the energy storage battery under test during charging. Finally, introduce the thermal resistance correction coefficient to correct the temperature balance, that is, take the product of the temperature decay rate and the thermal resistance correction coefficient as the thermal relaxation equilibrium index of the energy storage battery under test during charging.

[0037] In step 104, jointly predict the cycle life of the energy storage battery under test through the scale entropy health index and the thermal relaxation equilibrium index to obtain the joint prediction value of the remaining cycle life in the energy storage battery under test.

[0038] In some embodiments, the cycle life of the energy storage battery to be measured is jointly predicted by the scale entropy health index and the thermal relaxation balance index, and a joint prediction value of the remaining cycle life in the energy storage battery to be measured is obtained. Refer to Figure 2 As shown, this figure is a schematic flowchart of realizing joint prediction in some embodiments of the present application. The following steps can be adopted to realize joint prediction in this embodiment: In step 1041, a joint prediction model based on a boosting tree is initialized; In step 1042, the scale entropy health index is used as the voltage feature in the joint prediction model; In step 1043, the thermal relaxation balance index is used as the heat dissipation feature in the joint prediction model; In step 1044, the joint prediction model is used to predict the cycle life of the energy storage battery to be measured, and a joint prediction value of the remaining cycle life in the energy storage battery to be measured is obtained.

[0039] It should be noted that in the present application, the joint prediction value is a quantitative evaluation result of the remaining cycle life of the battery based on multi-dimensional feature fusion; the joint prediction model is a machine learning algorithm based on gradient boosting decision trees, which realizes multi-dimensional evaluation of battery life by fusing voltage features (scale entropy health index) and heat dissipation features (thermal relaxation balance index). The voltage feature quantifies the degradation degree of the internal electrochemical state of the battery through wavelet multi-scale entropy analysis, reflecting the loss of electrode material activity and the increase of interfacial impedance; the heat dissipation feature characterizes the battery thermal management efficiency through thermal relaxation dynamics analysis, capturing the decline of temperature regulation ability caused by the decline of the cooling system. When training the joint prediction model, a historical aging data set can be used, with the actual cycle life as the supervision target, and the decision tree is iteratively constructed to learn the non-linear mapping relationship between features and life. In the prediction stage, the output results of multiple trees are integrated, and at the same time, the point estimate and confidence interval of the remaining cycle times are given, with both prediction accuracy and reliability evaluation ability.

[0040] In some embodiments, the present application also provides a computer device, which includes a memory and a processor. The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-mentioned method for predicting the life of the energy storage battery in the industrial and commercial energy storage system.

[0041] In some embodiments, refer to Figure 3 , the dashed line in this figure indicates that the unit or the module is optional. This figure is a schematic structural diagram of a computer device for realizing the method for predicting the life of the energy storage battery in the industrial and commercial energy storage system according to the embodiments of the present application. The method for predicting the life of the energy storage battery in the industrial and commercial energy storage system described in the above embodiments can be passed through Figure 3It is implemented by the computer device shown. The computer device includes at least one processor 301, a memory 302, and at least one communication unit 305. The computer device can be a terminal device, a server, or a chip.

[0042] The processor 301 can be a general-purpose processor or a special-purpose processor. For example, the processor 301 can be a central processing unit (CPU). The CPU can be used to control the computer device, execute software programs, and process the data of software programs. The computer device can also include a communication unit 305 for implementing signal input (reception) and output (transmission).

[0043] For example, the computer device can be a chip. The communication unit 305 can be the input and / or output circuit of the chip, or the communication interface of the chip. The chip can be a component of a terminal device, a network device, or other devices.

[0044] Again, for example, the computer device can be a terminal device or a server. The communication unit 305 can be the transceiver of the terminal device or the server, or the transceiver circuit of the terminal device or the server.

[0045] The computer device can include one or more memories 302 with a program 304 stored thereon. The program 304 can be run by the processor 301 to generate instructions 303, enabling the processor 301 to execute the methods described in the above method embodiments according to the instructions 303. Optionally, data (such as a target audit model) can also be stored in the memory 302. Optionally, the processor 301 can also read the data stored in the memory 302. The data can be stored at the same storage address as the program 304, or at a different storage address from the program 304.

[0046] The processor 301 and the memory 302 can be set separately or integrated together. For example, they can be integrated on a system on chip (SOC) of a terminal device.

[0047] It should be understood that each step of the above method embodiments can be completed by a logic circuit in the form of hardware or an instruction in the form of software in the processor 301. The processor 301 can be a CPU, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices. For example, discrete gates, transistor logic devices, or discrete hardware components.

[0048] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0049] For example, in some embodiments, the present application also provides a computer-readable storage medium. Instructions or codes are stored in the computer-readable storage medium. When the instructions or codes are run on a computer, the computer is caused to execute the above-mentioned method for predicting the life of energy storage batteries in an industrial and commercial energy storage system.

[0050] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be construed as including the preferred embodiments as well as all changes and modifications that fall within the scope of the present application.

[0051] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these changes and modifications.

Claims

1. A method for predicting the life of energy storage batteries in an industrial and commercial energy storage system, characterized in that, The steps are as follows: Connect the energy storage battery to be measured in the industrial and commercial energy storage system to a standard charging device, and collect the voltage time series signal and battery temperature data of the energy storage battery to be measured during multiple charging processes; Perform linear interpolation on the voltage time series signal, and then extract the sample entropy and permutation entropy of the voltage components in the multi-scale space. Determine the battery aging sensitive scale in each scale space according to the sample entropy and permutation entropy of each voltage component. Based on the multi-scale voltage trajectory matrix of the energy storage battery to be measured during the charging process, perform a health assessment on the battery aging sensitive scale to obtain the scale entropy health index of the energy storage battery to be measured; Extract the surface maximum temperature of the energy storage battery to be measured at the end of the constant current stage and the equilibrium temperature after standing for a specified period of time from the battery temperature data, and then determine the thermal relaxation equilibrium index of the energy storage battery to be measured during the charging process based on the surface maximum temperature and the equilibrium temperature; Jointly predict the cycle life of the energy storage battery to be measured through the scale entropy health index and the thermal relaxation equilibrium index to obtain the joint prediction value of the remaining cycle life in the energy storage battery to be measured.

2. The method according to claim 1, wherein Performing linear interpolation on the voltage time series signal, and then extracting the sample entropy and permutation entropy of the voltage components in the multi-scale space specifically includes: Perform multi-layer decomposition on the voltage time series signal to obtain multiple scale spaces, and then use cubic spline interpolation on the voltage time series signal to the specified sampling rate; Calculate the voltage components of the specified time window in each scale space through the voltage time series signal after spline interpolation; Determine the sample entropy and permutation entropy of the voltage components in the multi-scale space through all voltage components.

3. The method according to claim 1, characterized in that, Determining the battery aging sensitive scale in each scale space according to the sample entropy and permutation entropy of each voltage component specifically includes: For each scale space, obtain the initial sample entropy and initial permutation entropy of the energy storage battery to be measured in the scale space; Determine the sample entropy decay rate of the voltage components in the scale space through the initial sample entropy and the sample entropy, and at the same time determine the permutation entropy decay rate of the voltage components in the scale space through the initial permutation entropy and the permutation entropy; Determine the battery aging sensitive scale in the scale space according to the sample entropy decay rate and the permutation entropy decay rate, and then obtain the battery aging sensitive scale in each scale space.

4. The method according to claim 1, wherein Performing a health assessment on the battery aging sensitive scale based on the multi-scale voltage trajectory matrix of the energy storage battery to be measured during the charging process to obtain the scale entropy health index of the energy storage battery to be measured specifically includes: Construct a voltage trajectory matrix containing sensitive scale components based on the voltage time series signal; Extract the variance contribution rate of the first principal component in the voltage trajectory matrix through principal component analysis as the health benchmark value; Determine the scale entropy health index of the energy storage battery to be measured according to the health benchmark value and the battery aging sensitive scale.

5. The method according to claim 1, characterized in that, Extracting the surface maximum temperature of the energy storage battery to be measured at the end of the constant current stage and the equilibrium temperature after standing for a specified period of time from the battery temperature data specifically includes: Obtain multiple surface temperatures of the energy storage battery to be measured at the end of the constant current stage and multiple standing temperatures after standing for a specified period of time from the battery temperature data; Select the maximum surface temperature of the energy storage battery under test at the end of the constant current stage from all surface temperatures; Determine the equilibrium temperature of the energy storage battery under test after a specified period of standing through all standing temperatures.

6. The method according to claim 1, wherein Determining the thermal relaxation equilibrium index of the energy storage battery under test during charging based on the maximum surface temperature and the equilibrium temperature specifically includes: Obtain the thermal resistance correction coefficient and the standing time of the energy storage battery under test; Determine the temperature decay rate of the energy storage battery under test during charging through the maximum surface temperature, the standing time, and the equilibrium temperature; Determine the thermal relaxation equilibrium index of the energy storage battery under test during charging according to the temperature decay rate and the thermal resistance correction coefficient.

7. The method according to claim 1, wherein Jointly predicting the cycle life of the energy storage battery under test through the scale entropy health index and the thermal relaxation equilibrium index to obtain the joint prediction value of the remaining cycle life in the energy storage battery under test specifically includes: Initialize an ensemble learning-based joint prediction model; Use the scale entropy health index as the voltage feature in the joint prediction model; Use the thermal relaxation equilibrium index as the heat dissipation feature in the joint prediction model; Use the joint prediction model to predict the cycle life of the energy storage battery under test to obtain the joint prediction value of the remaining cycle life in the energy storage battery under test.

8. The method according to claim 1, characterized in that, Use a high-precision data acquisition card to collect the voltage time series signal of the energy storage battery under test during multiple charging processes.

9. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the energy storage battery life prediction method in the industrial and commercial energy storage system according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, Instructions or codes are stored in the computer-readable storage medium. When the instructions or codes run on a computer, the computer is caused to execute the energy storage battery life prediction method in the industrial and commercial energy storage system according to any one of claims 1 to 8.

Images