Battery energy storage health state assessment method and system

Abstract

The invention relates to the technical field of battery energy storage, and discloses a battery energy storage health state assessment method and system, and the method comprises the steps: analyzing the constant-power charging and discharging process of a lithium battery, and obtaining health factors highly related to the battery energy storage health state through Pearson correlation analysis; constructing a bidirectional long short-term memory network model, taking the health factor highly related to the battery energy storage health state as model input, and outputting an energy storage battery health state estimation value; and optimizing the bidirectional long-short-term memory network model by improving a Kepler optimization algorithm and introducing a differential evolution strategy. According to the invention, the improved Kepler optimization algorithm is combined with the bidirectional long short-term memory network (BiLSTM) and the attention mechanism (Attention), so that the accurate estimation of the health state of the battery energy storage system is realized, the safety and reliability of the battery management system are improved, the high-precision evaluation of the health state of the lithium battery is realized, and the reliability of the battery management system is improved. And the robustness and generalization ability of the model under complex working conditions are improved.

Description

Technical Field

[0001] The present invention relates to the field of battery energy storage technology, and in particular to a battery energy storage health status assessment method and system. Background Art

[0002] With the rapid development of energy storage power stations, the health status assessment of battery energy storage systems has become an important technical means to ensure the stable operation and safe use of the system. Traditional methods are difficult to achieve efficient and accurate health status assessment due to unclear physical meanings, long time and huge aging test data, etc. Therefore, it is particularly urgent to develop a health status assessment method and system based on data-driven and advanced optimization algorithms. Summary of the invention

[0003] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0004] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a battery energy storage health status assessment method to solve the problem of how to achieve efficient and accurate health status assessment.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for evaluating the health status of battery energy storage, comprising: analyzing the constant power charging and discharging process of a lithium battery, extracting key characteristic indicators of multiple battery health factors, and obtaining health factors that are highly correlated with the health status of the battery energy storage through Pearson correlation analysis; Constructing a bidirectional long short-term memory network model, taking the health factor highly correlated with the health status of the battery energy storage as the model input, and outputting an estimated value of the health status of the energy storage battery; By improving the Kepler optimization algorithm and introducing a differential evolution strategy, the bidirectional long short-term memory network model is optimized to obtain the optimal model configuration.

[0006] As a preferred solution of the battery energy storage health status assessment method of the present invention, obtaining a health factor highly correlated with the battery energy storage health status includes: Based on the analysis of the constant power charging and discharging process of lithium batteries, multiple health factors are extracted, including constant current charging time, constant voltage charging time, constant current discharging time, constant voltage discharging time, peak temperature during charging, and peak temperature during discharging; The correlation coefficient between the health factor and the battery health status is calculated through Pearson correlation analysis, and the absolute value of the correlation coefficient is compared with a preset coefficient threshold range to screen out the health factor that is highly correlated with the battery health status.

[0007] As a preferred solution of the battery energy storage health status assessment method described in the present invention, wherein: the construction of a bidirectional long short-term memory network model includes: The bidirectional long short-term memory network model includes two parallel LSTM layers in opposite directions, namely a forward LSTM layer and a backward LSTM layer; Calculation is performed through the bidirectional long short-term memory network model, and past and future context information processing is used to aggregate the outputs of the LSTM layers in two directions; The forward LSTM layer performs information processing along the forward direction of the time series, and the backward LSTM layer performs backward information processing in reverse time sequence.

[0008] As a preferred solution of the battery energy storage health status assessment method described in the present invention, the calculation of the bidirectional long short-term memory network model includes: The forward LSTM layer reads data sequentially from the start point to the end point of the sequence, gradually accumulates and transmits the forward hidden state to capture the historical information up to the current moment; The backward LSTM layer traces back from the end point of the sequence to the starting point, accumulating and transmitting the backward hidden state in a point-by-point updating manner to capture relevant information from future time points; The hidden states in both directions are merged to generate a comprehensive representation that contains the full temporal context information.

[0009] As a preferred solution of the battery energy storage health status assessment method described in the present invention, the calculation formula of the bidirectional long short-term memory network model is expressed as: , in, represents the input of the model at time t, represents the forward LSTM layer weight matrix, represents the backward LSTM layer weight matrix, represents the output at time t-1, represents the output at time t, represents the output at time t+1, represents the output at time t, represents the bias vector, Represents the final output of the model.

[0010] As a preferred solution of the battery energy storage health status assessment method described in the present invention, the improved Kepler optimization algorithm includes: The initialization of the good point set is expressed as: , in, For the best point collection, is the best point coefficient, i is the best point, which can be expressed as: , Among them, P is satisfied The smallest prime number of Represents the dimension of the problem to be optimized; By initializing the good point set, the Kepler optimization algorithm is improved and the data initialization strategy of key characteristic indicators of battery energy storage is established, which is expressed as: , in, It represents the position of the i-th planet individual in the solution space in the j-th dimension, that is, the position of the key characteristic indicator data of battery energy storage in the search space, N represents the population size, , Respectively represent the lower and upper bounds of data search.

[0011] As a preferred solution of the battery energy storage health status assessment method described in the present invention, the differential evolution strategy includes: The Kepler optimization algorithm uses the differential mutation operation in the differential evolution strategy to generate new candidate solutions, and the candidate solutions are combined with individuals in the current population through the differential crossover operation; The differential mutation operation generates a mutation vector through the differences between different individuals in the current population, and the mutation vector guides the movement direction or adjustment position of the planetary individual; The differential crossover operation incorporates the beneficial information in the mutation vector into the existing individuals to generate a diversified solution set.

[0012] In a second aspect, the present invention provides a battery energy storage health status assessment system, comprising: The analysis module is used to analyze the constant power charging and discharging process of lithium batteries, extract key characteristic indicators of multiple battery health factors, and obtain health factors that are highly correlated with the health status of battery energy storage through Pearson correlation analysis; A construction module is used to construct a bidirectional long short-term memory network model, using the health factor highly correlated with the health status of the battery energy storage as the model input, and outputting an estimated value of the health status of the energy storage battery; The optimization module is used to optimize the bidirectional long short-term memory network model by improving the Kepler optimization algorithm and introducing a differential evolution strategy to obtain an optimal model configuration.

[0013] In a third aspect, the present invention provides an electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the battery energy storage health status assessment method are implemented.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the battery energy storage health status assessment method.

[0015] Compared with the prior art, the present invention has the following beneficial effects: the present invention utilizes an improved Kepler optimization algorithm combined with a bidirectional long short-term memory network (BiLSTM) and an attention mechanism (Attention) to achieve accurate estimation of the health status of the battery energy storage system, thereby improving the safety and reliability of the battery management system; combining multiple health factors and advanced data-driven models to achieve high-precision evaluation of the health status of lithium batteries, and by improving the Kepler optimization algorithm, improves the robustness and generalization ability of the model under complex working conditions, and the model training and optimization process is fast, meeting the needs of real-time health status monitoring, and has a simple structure, which is convenient for promotion and application in energy storage systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0017] Figure 1 A schematic diagram of the overall process of a method for evaluating the health status of battery energy storage according to an embodiment of the present invention; Figure 2 It is a schematic diagram of the training set effect before and after the improvement of the battery energy storage health status assessment method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0019] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0020] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0021] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0022] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0023] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0024] Example 1, reference Figure 1 Table 1 is an embodiment of the present invention, which provides a method for evaluating the health status of battery energy storage, including: S101, analyzing the constant power charging and discharging process of the lithium battery, extracting the key characteristic indicators of multiple battery health factors, and obtaining the health factors that are highly correlated with the health status of the battery energy storage through Pearson correlation analysis; S102, constructing a bidirectional long short-term memory network model, taking a health factor highly correlated with the health status of the battery energy storage as a model input, and outputting an estimated value of the health status of the energy storage battery; S103, by improving the Kepler optimization algorithm and introducing the differential evolution strategy, the bidirectional long short-term memory network model is optimized to obtain the optimal model configuration.

[0025] It should be noted that the present invention achieves high-precision assessment of the health status of lithium batteries by combining a bidirectional long short-term memory network (BiLSTM) and an attention mechanism, combining multiple health factors and advanced data-driven models, and introducing the Kepler optimization algorithm (KOA-DE) improved by a differential evolution strategy.

[0026] In step S101, preferably, obtaining a health factor highly related to the health status of the battery energy storage includes: Based on the analysis of the constant power charging and discharging process of lithium batteries, multiple health factors are extracted, including constant current charging time, constant voltage charging time, constant current discharging time, constant voltage discharging time, peak temperature during charging, and peak temperature during discharging; Through Pearson correlation analysis, the correlation coefficient between the health factor and the battery health status is calculated, and the absolute value of the correlation coefficient is compared with the preset coefficient threshold range to screen out the health factor that is highly correlated with the battery health status.

[0027] For example, when evaluating the state of health (SOH) of a battery energy storage system, this embodiment selects six key characteristic indicators: constant current charging time, constant voltage charging time, constant current discharging time, constant voltage discharging time, peak temperature during charging, and peak temperature during discharging. In order to deeply explore the intrinsic connection and sensitivity between these health factors and the battery SOH, the Pearson correlation analysis method is used to quantify the strength of the correlation between them.

[0028] The Pearson correlation coefficient, whose range is limited to between -1 and 1, is a statistical indicator for evaluating the strength and direction of the linear relationship between two variables. The closer the value is to 1, the stronger the positive correlation is between the two variables; the closer it is to -1, the stronger the negative correlation is between them. The absolute value of the coefficient directly reflects the strength of the correlation.

[0029] Specifically, the absolute value of the Pearson correlation coefficient in the range of 0.2 to 0.4 is considered to be a low to medium correlation, 0.4 to 0.6 indicates a medium correlation, and when the absolute value exceeds 0.6, it is considered to be a high correlation. In particular, when the coefficient value is greater than 0.7, it indicates that there is an extremely strong correlation between the health factor and the battery SOH.

[0030] Table 1 summarizes the results of the Pearson correlation analysis between the six health factors and the battery SOH, revealing the individual contribution of each health factor to the SOH assessment, and providing basic data support for building a more accurate and effective battery health status assessment model. By comprehensively considering these highly correlated health factors, it is possible to fully capture the signs of battery performance degradation, thereby accurately monitoring and predicting the battery's SOH. In Table 1, the peak charging temperature is moderately correlated with the battery energy storage capacity, and other key characteristic indicators are strongly correlated with the battery energy storage capacity.

[0031] Table 1: Correlation of key characteristic indicators of different battery energy storage

[0032] In step S102, preferably, constructing a bidirectional long short-term memory network model includes: The bidirectional long short-term memory network model includes two parallel LSTM layers in opposite directions, namely the forward LSTM layer and the backward LSTM layer; The calculation is performed through a bidirectional long short-term memory network model, and the past and future context information are processed at the same time to aggregate the outputs of the LSTM layers in two directions; Among them, the forward LSTM layer processes information along the forward direction of the time series, and the backward LSTM layer performs backward information processing in the reverse time sequence.

[0033] It should be noted that the bidirectional long short-term memory network (Bi-directional LongShort-Term Memory, BiLSTM) adopted in this embodiment is an advanced extension of the long short-term memory network (LSTM). Its uniqueness lies in the integration of two parallel and opposite LSTM layers. Its core advantage is that it can simultaneously utilize past and future contextual information, and by aggregating the outputs of the LSTM layers in these two directions, it can achieve comprehensive capture and in-depth analysis of bidirectional dependencies in data sequences.

[0034] Preferably, the calculation of the bidirectional long short-term memory network model includes: The forward LSTM layer reads data sequentially from the start point to the end point of the sequence, gradually accumulating and passing forward hidden states to capture historical information up to the current moment; The backward LSTM layer traces back from the end point of the sequence to the starting point, accumulating and passing the backward hidden state in a point-by-point update manner to capture relevant information from future time points; The hidden states in both directions are merged to generate a comprehensive representation that contains the full temporal context information.

[0035] It should be noted that the hidden states in the two directions are merged, usually through a simple concatenation operation, to generate a comprehensive representation containing complete temporal context information. This process not only enhances the model's ability to understand sequence data, but also significantly improves its performance in processing complex tasks that require consideration of both contextual information.

[0036] Preferably, the calculation formula of the bidirectional long short-term memory network model is expressed as: , in, represents the input of the model at time t, i.e., the key characteristic indicators of different battery energy storage in Table 1, and the output is the estimated SOH value of the energy storage battery. represents the forward LSTM layer weight matrix, represents the backward LSTM layer weight matrix, represents the output at time t-1, represents the output at time t, represents the output at time t+1, represents the output at time t, represents the bias vector, Represents the final output of the model.

[0037] In step S103, it should be noted that the Kepler Optimization Algorithm (KOA), as an innovative meta-heuristic optimization strategy based on physical principles, is deeply inspired by Kepler's laws of planetary motion and aims to simulate the laws of celestial motion to predict and optimize parameters in complex systems. Although KOA has shown excellent adaptability and broad application potential in dealing with nonlinear optimization problems, its performance is highly dependent on the careful selection of the initial solution set and parameters. In addition, like the common challenges faced by many optimization algorithms, KOA also has the risk of falling into the local optimal trap, which may miss the global optimal solution, especially when facing highly complex problems, its performance may be significantly limited.

[0038] In view of the inherent uncertainty and uneven spatial distribution of the random generation of the initial population of the KOA algorithm, this embodiment proposes a more sophisticated initialization strategy, namely the good point set initialization method. This method aims to construct a set of initial solutions that are uniformly distributed and representative in space through mathematical means, thereby effectively reducing the algorithm's dependence on the sensitivity of initial conditions.

[0039] Specifically, the good point set initialization uses low-discrepancy sequences or specific mathematical rules to ensure that the generated initial population can evenly cover the solution space, which not only retains the random exploration ability but also enhances the algorithm's search efficiency on a global scale. Through this improvement, the KOA algorithm can obtain a more balanced and representative starting point at the startup stage, laying a solid foundation for the subsequent optimization process.

[0040] The good point set is initialized as follows: , in, For the best point collection, is the best point coefficient, i For the best point, It can be expressed as: , Among them, P is satisfied The smallest prime number of Represents the dimension of the problem to be optimized.

[0041] Preferably, the Kepler optimization algorithm is improved by initializing the good point set, and a data initialization strategy for key characteristic indicators of battery energy storage is established, which is expressed as: , in, It represents the position of the i-th planet individual in the solution space in the j-th dimension, that is, the position of the key characteristic indicator data of battery energy storage in the search space, N represents the population size, , Respectively represent the lower and upper bounds of data search.

[0042] It should be noted that in the process of solving the optimization problem, KOA updates the individual positions in the solution space through continuous exploration and development, and adopts an elite retention strategy to retain the best individuals in each generation. The elite retention mechanism is an effective greedy strategy, which ensures that the algorithm will at least not lose the known best solution during the iteration process. However, this strategy also potentially introduces the risk of local optimal traps, especially when the retained optimal individual only represents a local optimal solution.

[0043] In order to overcome this limitation, improve the global search capability of the algorithm and accelerate the process of convergence to the global optimal solution, this embodiment incorporates the differential evolution (DE) strategy into the KOA algorithm. Differential evolution is an optimization algorithm based on population evolution, which searches for the optimal solution by simulating the principles of biological evolution in nature.

[0044] Preferably, the differential evolution strategy includes: The Kepler optimization algorithm uses the differential mutation operation in the differential evolution strategy to generate new candidate solutions, which are combined with individuals in the current population through differential crossover operations; Among them, the differential mutation operation generates mutation vectors through the differences between different individuals in the current population. The mutation vectors guide the movement direction or adjustment position of the planetary individuals. The differential crossover operation incorporates the beneficial information in the mutation vector into the existing individuals to generate a diverse solution set.

[0045] Exemplarily, in this hybrid strategy, KOA uses the differential mutation operation in differential evolution to generate new candidate solutions, which is expressed as: , in, For the individual after mutation, are the different individuals in the population, and F is the scaling factor.

[0046] These candidate solutions are combined with individuals in the current population through a differential crossover operation to produce a more diverse solution set, expressed as: , in, is the new population, and CR is the population crossover probability.

[0047] It should be noted that the differential mutation operation generates a mutation vector through the differences between different individuals in the current population, which is then used to guide the movement direction of the planetary individual or adjust its position. The differential crossover operation integrates the beneficial information in the mutation vector into the existing individuals through a specific crossover mechanism to form a new candidate solution. In this way, the algorithm can effectively jump out of the local optimal solution while maintaining a certain degree of local search capability and explore a wider solution space, thereby improving the global search capability and robustness of the algorithm. By integrating the differential evolution strategy into the KOA algorithm, it can not only alleviate the local optimal problem that may be caused by elite retention, but also significantly improve the algorithm's global search efficiency and solution quality in complex optimization problems.

[0048] The present invention utilizes an improved Kepler optimization algorithm combined with a bidirectional long short-term memory network (BiLSTM) and an attention mechanism to achieve accurate estimation of the health status of the battery energy storage system and improve the safety and reliability of the battery management system. It combines multiple health factors and advanced data-driven models to achieve high-precision evaluation of the health status of lithium batteries. By improving the Kepler optimization algorithm, the robustness and generalization ability of the model under complex working conditions are improved. The model training and optimization process are fast, meeting the needs of real-time health status monitoring. The structure is simple and easy to promote and apply in energy storage systems.

[0049] The above is a schematic scheme of a battery energy storage health status assessment method of this embodiment. It should be noted that the technical scheme of the battery energy storage health status assessment system and the technical scheme of the above-mentioned battery energy storage health status assessment method belong to the same concept, and the details of the technical scheme of the battery energy storage health status assessment system not described in detail in this embodiment can be referred to the description of the technical scheme of the above-mentioned battery energy storage health status assessment method.

[0050] The battery energy storage health status assessment system in this embodiment includes: The analysis module is used to analyze the constant power charging and discharging process of lithium batteries, extract key characteristic indicators of multiple battery health factors, and obtain health factors that are highly correlated with the health status of battery energy storage through Pearson correlation analysis; A construction module is used to construct a bidirectional long short-term memory network model, which uses a health factor highly correlated with the health status of the battery energy storage as the model input and outputs an estimated value of the health status of the energy storage battery; The optimization module is used to optimize the bidirectional long short-term memory network model by improving the Kepler optimization algorithm and introducing the differential evolution strategy to obtain the optimal model configuration.

[0051] This embodiment also provides an electronic device, which is applicable to battery energy storage health status assessment, including: Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the battery energy storage health status assessment method proposed in the above embodiment.

[0052] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for evaluating the health status of battery energy storage proposed in the above embodiment is implemented.

[0053] The storage medium proposed in this embodiment and the method for implementing the battery energy storage health status assessment proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0054] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0055] Example 2, reference Figure 2 And Table 2 is an embodiment of the present invention, which provides a battery energy storage health status assessment method. In order to verify its beneficial effects, the comparison results of two schemes are provided.

[0056] In order to verify the accuracy of the estimation model proposed in the present invention in the prediction of lithium battery energy storage, an improved whale optimization algorithm-convolutional neural network-long short-term memory network deep learning (KOA-BiLSTM-Attention) model was constructed. In the data processing stage, the lithium battery dataset publicly available on the Internet was used, and abnormal data was eliminated through preprocessing steps to ensure the quality and reliability of the input data. Subsequently, the dataset was divided into training set and test set in a ratio of 70% and 30% to ensure the independence of model training and verification.

[0057] During the model training and optimization process, the experimental parameters were set: the maximum number of iterations was 1000, the population size was 30, the maximum number of function evaluations was 30, the batch size was 30, the initial learning rate was 0.01, the learning rate reduction factor was 0.1, and the lower and upper bounds of the data search were set to 0.2 and 0.9, respectively.

[0058] By comparing the performance of the Whale Optimization Algorithm-Bidirectional Long Short-Term Memory Network-Attention Mechanism (WOA-CNN-LSTM) model and the Whale Optimization Algorithm-Convolutional Neural Network-Long Short-Term Memory Network Deep Learning (KOA-BiLSTM-Attention) model on the training set, Figure 2As shown, it can be observed that the improved model has higher prediction accuracy in multiple time intervals. Specifically, in the time period of 0-50, the prediction error of the battery energy storage SOH model of KOA-BiLSTM-Attention is smaller than before the improvement; and in the time period of 90-118, the battery energy storage SOH prediction curve of KOA-BiLSTM-Attention is closer to the true value. These results show that the improved model of the present invention has significant advantages and potential in the field of lithium battery energy storage prediction.

[0059] To evaluate the performance of the improved model, the root mean square error (RMSE), mean absolute percentage error (MAPE) and mean absolute error (MAE) were used as evaluation indicators.

[0060] Table 2: Model performance evaluation

[0061] Table 2 shows the RMSE, MAPE, and MAE results of different data-driven models. The KOA-BiLSTM-Attention model proposed in the present invention is 2.8% lower in RMSE than the WOA-CNN-LSTM model.

[0062] In summary, as the battery energy storage enters the aging stage, the KOA-BiLSTM-Attention model estimation model has a more accurate estimation result for SOH.

[0063] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for evaluating the health status of battery energy storage, characterized in that: include: Analyze the constant power charging and discharging process of lithium batteries, extract key characteristic indicators of multiple battery health factors, and obtain health factors that are highly correlated with the health status of battery energy storage through Pearson correlation analysis; Constructing a bidirectional long short-term memory network model, taking the health factor highly correlated with the health status of the battery energy storage as the model input, and outputting an estimated value of the health status of the energy storage battery; By improving the Kepler optimization algorithm and introducing a differential evolution strategy, the bidirectional long short-term memory network model is optimized to obtain the optimal model configuration.

2. The battery energy storage health status assessment method according to claim 1, characterized in that: The health factors that are highly correlated with the health status of battery energy storage include: Based on the analysis of the constant power charging and discharging process of lithium batteries, multiple health factors are extracted, including constant current charging time, constant voltage charging time, constant current discharging time, constant voltage discharging time, peak temperature during charging, and peak temperature during discharging; The correlation coefficient between the health factor and the battery health status is calculated through Pearson correlation analysis, and the absolute value of the correlation coefficient is compared with a preset coefficient threshold range to screen out the health factor that is highly correlated with the battery health status.

3. The battery energy storage health status assessment method according to claim 1, characterized in that: The construction of the bidirectional long short-term memory network model includes: The bidirectional long short-term memory network model includes two parallel LSTM layers in opposite directions, namely a forward LSTM layer and a backward LSTM layer; Calculation is performed through the bidirectional long short-term memory network model, and past and future context information processing is used to aggregate the outputs of the LSTM layers in two directions; The forward LSTM layer performs information processing along the forward direction of the time series, and the backward LSTM layer performs backward information processing in reverse time sequence.

4. The battery energy storage health status assessment method according to claim 3, characterized in that: The calculation of the bidirectional long short-term memory network model includes: The forward LSTM layer reads data sequentially from the start point to the end point of the sequence, gradually accumulates and transmits the forward hidden state to capture the historical information up to the current moment; The backward LSTM layer traces back from the end point of the sequence to the starting point, accumulating and transmitting the backward hidden state in a point-by-point updating manner to capture relevant information from future time points; The hidden states in both directions are merged to generate a comprehensive representation that contains the full temporal context information.

5. The battery energy storage health status assessment method according to claim 4, characterized in that: The calculation formula of the bidirectional long short-term memory network model is expressed as: ， in, represents the input of the model at time t, represents the forward LSTM layer weight matrix, represents the backward LSTM layer weight matrix, represents the output at time t-1, represents the output at time t, represents the output at time t+1, represents the output at time t, represents the bias vector, Represents the final output of the model.

6. The battery energy storage health status assessment method according to claim 5, characterized in that: The improved Kepler optimization algorithm comprises: The initialization of the good point set is expressed as: ， in, For the best point collection, is the best point coefficient, i is the best point, which can be expressed as: ， Among them, P is satisfied The smallest prime number of Represents the dimension of the problem to be optimized; By initializing the good point set, the Kepler optimization algorithm is improved and the data initialization strategy of key characteristic indicators of battery energy storage is established, which is expressed as: ， in, It represents the position of the i-th planet individual in the solution space in the j-th dimension, that is, the position of the key characteristic indicator data of battery energy storage in the search space, N represents the population size, , Respectively represent the lower and upper bounds of data search.

7. The battery energy storage health status assessment method according to any one of claims 4 to 6, characterized in that: The differential evolution strategy includes: The Kepler optimization algorithm uses the differential mutation operation in the differential evolution strategy to generate new candidate solutions, and the candidate solutions are combined with individuals in the current population through the differential crossover operation; The differential mutation operation generates a mutation vector through the differences between different individuals in the current population, and the mutation vector guides the movement direction or adjustment position of the planetary individual; The differential crossover operation incorporates the beneficial information in the mutation vector into the existing individuals to generate a diversified solution set.

8. A battery energy storage health status assessment system, characterized in that: include, The analysis module is used to analyze the constant power charging and discharging process of lithium batteries, extract the key characteristic indicators of multiple battery health factors, and obtain the health factors that are highly correlated with the health status of battery energy storage through Pearson correlation analysis; A construction module is used to construct a bidirectional long short-term memory network model, using the health factor highly correlated with the health status of the battery energy storage as a model input, and outputting an estimated value of the health status of the energy storage battery; The optimization module is used to optimize the bidirectional long short-term memory network model by improving the Kepler optimization algorithm and introducing a differential evolution strategy to obtain an optimal model configuration.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the battery energy storage health status assessment method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: It stores computer executable instructions, which, when executed by a processor, can implement the steps of the battery energy storage health status assessment method described in any one of claims 1 to 7.

Images