Method, device and equipment for determining state of charge of battery and readable storage medium

Abstract

The invention discloses a method, device and equipment for determining the state of charge of a battery and a readable storage medium, and relates to the technical field of battery management, and the method comprises the steps: building a time kernel function and a space kernel function according to the historical state parameters of a single battery, and building a global integration model according to the time kernel function and the space kernel function, and finally, according to the measured value of the real-time charge state of the target battery monomer and the global integration model, determining the charge state of the battery monomer. The correlation of the battery monomers in time evolution and space distribution is comprehensively described through the time kernel function and the space kernel function, and then the global integration model with global coupling characteristics is constructed, so that the accuracy of the state of charge determined according to the global integration model is improved.

Description

Technical Field

[0001] The present application relates to the technical field of battery management, and particularly to a method, device, equipment, and readable storage medium for determining the state of charge of a battery. Background Art

[0002] In a battery management system, the state of charge (SOC) is a fundamental parameter for evaluating battery performance and an important support for realizing the full life cycle management of the battery.

[0003] Currently, SOC is usually estimated by direct measurement or by establishing an equivalent model. Although the direct measurement method is simple to implement, it is easily affected by inaccurate initial values and the accumulation of sensor errors, resulting in the estimated result gradually deviating from the actual value. In contrast, the method of establishing an equivalent model can usually obtain higher estimation accuracy, but there is often a problem that the model has a strong dependence on the application scenario, that is, the generalization ability of the model is poor, which in turn limits the accuracy and applicability of SOC estimation. Summary of the Invention

[0004] The present application provides a method, device, equipment, and readable storage medium for determining the state of charge of a battery, so as to at least solve the problem that the accuracy of determining the state of charge in the related art is limited.

[0005] The present application provides a method for determining the state of charge of a battery, including: Obtaining historical state parameters of each battery cell of a target battery; the target battery includes a plurality of battery cells; Constructing a time kernel function and a space kernel function of the target battery according to the historical state parameters; the time kernel function is used to characterize the time correlation between the historical state parameters of the plurality of battery cells; the space kernel function is used to characterize the space correlation between the plurality of battery cells; Constructing a global integration model according to the time kernel function and the space kernel function; Performing a hidden state recursive estimation on the global integration model according to the measured value of the real-time state of charge of a target battery cell in the target battery, and determining an estimated value of the state of charge of the target battery cell; Determining the state of charge of the target battery cell according to the historical state parameters and the estimated value.

[0006] The present application further provides a device for determining the state of charge of a battery, including: an acquisition module, configured to acquire historical state parameters of each battery cell of a target battery; the target battery includes a plurality of battery cells; A construction module for constructing a time kernel function and a space kernel function of a target battery according to historical state parameters; the time kernel function is used to characterize the time correlation between the historical state parameters of multiple battery cells; the space kernel function is used to characterize the space correlation between multiple battery cells; The construction module is further configured to construct a global integration model according to the time kernel function and the space kernel function; A processing module for performing a hidden state recursive estimation on the global integration model according to the measured value of the real-time state of charge of a target battery cell in the target battery, and determining an estimated value of the state of charge of the target battery cell; A determination module for determining the state of charge of the target battery cell according to the historical state parameters and the estimated value.

[0007] This application also provides an electronic device, including: a memory for storing a computer program; a processor for implementing the steps of the method for determining the state of charge of the battery as described above when executing the computer program.

[0008] This application also provides a computer-readable storage medium, in which a computer program is stored, and wherein the computer program implements the steps of the method for determining the state of charge of the battery as described above when executed by a processor.

[0009] This application also provides a computer program product, including a computer program, and the computer program implements the steps of the method for determining the state of charge of the battery as described above when executed by a processor.

[0010] Through this application, since a time kernel function and a space kernel function can be established according to the historical state parameters of the battery cell, and a global integration model can be constructed according to the time kernel function and the space kernel function, and finally the state of charge of the battery cell can be determined according to the measured value of the real-time state of charge of the target battery cell and the global integration model. In this way, the time kernel function and the space kernel function comprehensively characterize the correlation of the battery cell in time evolution and space distribution, and then a global integration model with global coupling characteristics is constructed, thereby improving the accuracy of the state of charge determined according to the global integration model. In addition, it is also possible to robustly recursively estimate the state of charge of the battery cell in the presence of observation noise and initial uncertainty, further improving the accuracy and adaptability of the determined state of charge. Description of the Drawings

[0011] In order to more clearly illustrate the embodiments of the present application, the following will briefly introduce the drawings required for the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0012] Figure 1Schematic diagram of an application scenario of a method for determining the state of charge of a battery provided by an embodiment of the present application; Figure 2 One of the flowcharts of the method for determining the state of charge of a battery provided by an embodiment of the present application; Figure 3 Another flowchart of the method for determining the state of charge of a battery provided by an embodiment of the present application; Figure 4 Another flowchart of the method for determining the state of charge of a battery provided by an embodiment of the present application; Figure 5 Another flowchart of the method for determining the state of charge of a battery provided by an embodiment of the present application; Figure 6 Another flowchart of the method for determining the state of charge of a battery provided by an embodiment of the present application; Figure 7 Another flowchart of the method for determining the state of charge of a battery provided by an embodiment of the present application; Figure 8 Schematic diagram of the structure of a device for determining the state of charge of a battery provided by an embodiment of the present application. Detailed implementation manners

[0013] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present application.

[0014] It should be noted that in the description of the present application, the terms "include", "comprise" or any other variant thereof are intended to cover a non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article or device. The terms "first", "second", etc. in the present application are used to distinguish similar objects and not to describe a specific order or sequence.

[0015] To enable those skilled in the art of the present technology to better understand the solution of the present application, the present application will be further described in detail below in conjunction with the accompanying drawings and specific implementation manners.

[0016] Constructing a battery state of charge (SOC) estimation system is of great value for realizing the dynamic estimation of the battery working state. On the one hand, it can improve the energy utilization efficiency by optimizing the setting of charge and discharge thresholds. On the other hand, it can analyze the aging characteristics and establish a safety monitoring and early warning mechanism. Therefore, accurately, real-time, and effectively estimating the battery state of charge is not only directly related to the improvement of the operation efficiency of the battery energy management system but also has a decisive impact on promoting the breakthrough of the battery's endurance ability and the upgrade of its safety performance.

[0017] Currently, the state of charge is usually estimated by direct measurement, equivalent model-based, data-driven, etc. Among them, the ampere-hour integration method and the power integration method are direct measurement methods for estimating the state of charge, but they are easily affected by inaccurate initial values, and the sensor errors will gradually accumulate, resulting in an increase in the estimation error.

[0018] For the equivalent model-based methods, for example, the open-circuit voltage method estimates the state of charge equivalently through the relationship curve between the open-circuit voltage and the battery state of charge. However, this method requires a long time to statically place the battery and cannot estimate the battery state of charge under real-time working conditions. Another example is that the electrochemistry-based model method is to establish a model of the internal chemical reaction of the battery to explain the aging mechanism of the battery. However, the electrochemistry model usually contains multiple complex equations and highly coupled model parameters and is mainly applied to the research of the internal electrochemical reactions of the battery. The basic principles of the equivalent circuit model (ECM) and the fractional-order model (FOM) are to establish the relationship between the characteristic parameters and the state of charge through filtering algorithms by using the characteristic parameters shown by lithium batteries during cycling, and further estimate the state of charge of lithium batteries. It can be seen that although the equivalent model-based estimation method can estimate the battery state of charge relatively accurately, it is overly dependent on the accuracy of the model and parameters, and different models need to be established to deal with different situations, with weak generalization ability, thus limiting the accuracy and applicability of SOC estimation.

[0019] For the data-driven methods, for example, artificial neural network (ANN), support vector machine (SVM), etc., this method needs to be trained with a large amount of battery data to learn the non-linear mapping relationship between the input and output, so as to realize the estimation of the battery state of charge. Although these methods have good adaptability and generalization ability, they require a large amount of historical data for training, and the accuracy and reliability of the model largely depend on the quality and quantity of the data.

[0020] As can be seen from the above, the existing methods for estimating the state of charge of a battery have their own advantages and disadvantages. The direct measurement method is simple and easy to implement, but the problem of error accumulation is serious; the method based on an equivalent model can deeply characterize the internal characteristics of the battery, but it is difficult to balance the model complexity and generalization ability; the data-driven method has strong adaptability, but its dependence on data limits its application in some scenarios.

[0021] In view of the above problems, the embodiments of the present application provide a method, device, equipment, and readable storage medium for determining the state of charge of a battery. The method includes establishing a time kernel function and a space kernel function according to the historical state parameters of a battery cell, constructing a global integration model according to the time kernel function and the space kernel function, and finally determining the state of charge of the battery cell according to the measured value of the real-time state of charge of the target battery cell and the global integration model. In this way, the time kernel function and the space kernel function comprehensively characterize the correlation of the battery cell in time evolution and space distribution, and then a global integration model with global coupling characteristics is constructed, thereby improving the accuracy of the state of charge determined according to the global integration model.

[0022] Figure 1 FIG. is a schematic diagram of an application scenario of a method for determining the state of charge of a battery provided by an embodiment of the present application. As Figure 1 shown, this scenario includes a server 100 and an electronic device 200.

[0023] The electronic device 200 provided by the embodiments of the present application can have various implementation forms. For example, it can be a mobile phone, a personal computer (PC), a wearable device, a vehicle-mounted device, etc.

[0024] In some embodiments, when the electronic device 200 receives a determination instruction for the state of charge of the battery, it can perform data communication with the server 100. The electronic device 200 is allowed to communicate with the server 100 through a local area network (LAN) or a wireless local area network (WLAN).

[0025] The server 100 can be a server that provides various services. For example, it is a server that provides support for the historical state parameters uploaded by the electronic device 200. The server 100 can be a server cluster or multiple server clusters, and can include one type or multiple types of servers.

[0026] It should be noted that the method for determining the state of charge of the battery provided by the embodiments of the present application can be executed by the electronic device 200, or by the server 100, or jointly by the server 100 and the electronic device 200.

[0027] The device for determining the state of charge of a battery provided by an embodiment of the present application may be hardware or software. When the device for determining the state of charge of a battery is hardware, it may be various electronic devices 200 with the function of determining the state of charge of a battery, including but not limited to smart phones, TVs, tablet computers, smart watches, computers, AI devices, robots, smart vehicles, and so on. When the device for determining the state of charge of a battery is software, it may be installed in the above-listed electronic devices 200. It may be implemented as multiple software or software modules (for example, used to provide the service of determining the state of charge of a battery), or may be implemented as a single software or software module. No specific limitation is made here.

[0028] Figure 2 It is a schematic flow chart of a method for determining the state of charge of a battery provided by an embodiment of the present application. As Figure 2 shown, the method for determining the state of charge of a battery may include the following steps: S11. Obtain the historical state parameters of each battery cell of the target battery.

[0029] Among them, the target battery includes multiple battery cells. In some embodiments, the target battery is a lithium battery. The historical state parameters at least include the sampling time, electrical parameters, and the spatial position of the battery cell in the target battery (the spatial position can be represented by spatial coordinate points). The electrical parameters may include various state parameters related to the battery cell, such as voltage, current, temperature, state of charge, and so on.

[0030] Specifically, the manner of obtaining the historical state parameters of each battery cell of the target battery may be to obtain the historical state data of each battery cell of the target battery in the charging and discharging states under different working conditions.

[0031] S12. Construct a time kernel function and a spatial kernel function of the target battery according to the historical state parameters.

[0032] Among them, the time kernel function is used to characterize the time correlation between the historical state parameters of multiple battery cells; the spatial kernel function is used to characterize the spatial correlation between multiple battery cells.

[0033] Specifically, the method of constructing the time kernel function and the spatial kernel function of the target battery based on historical state parameters can be to construct the time kernel function of the target battery according to the sampling time and electrical parameters, and construct the spatial kernel function of the target battery according to the spatial position and electrical parameters. That is to say, the method of constructing the time kernel function of the target battery based on historical state parameters can be based on the sampling time, and by analyzing the time correlation between the historical state parameters of multiple battery cells at different time points, a time kernel function reflecting the law of state evolution over time is established. The method of constructing the spatial kernel function of the target battery based on historical state parameters can be based on the spatial position of the battery cell in the target battery, and by analyzing the spatial correlation between the historical state parameters of different cells, a spatial kernel function reflecting the spatial dependence relationship between cells is established. This spatial kernel function can express the similarity of the states of adjacent cells and the spatial structure characteristics, so as to assist in capturing the overall spatio-temporal dynamic characteristics of the target battery. In the above solution, constructing the time kernel function based on the sampling time and electrical parameters helps to reveal the law of battery performance change over time; using the spatial position and electrical parameters of the battery cell in the target battery to construct the spatial kernel function can reflect the differences and spatial distribution characteristics between cells at different positions. In this way, the operating state of the target battery is comprehensively characterized from both the time and space dimensions, improving the accuracy of state modeling and providing a more reliable data basis for subsequent state of charge assessment.

[0034] S13. Construct a global integration model according to the time kernel function and the spatial kernel function.

[0035] Among them, the global integration model is used to describe the evolution process of the state parameters of the battery cells in the target battery at different time points and spatial positions.

[0036] Specifically, the method of constructing the global integration model according to the time kernel function and the spatial kernel function can be to combine the time kernel function and the spatial kernel function to form a joint spatio-temporal kernel function, so as to construct a global integration model that simultaneously considers time and space dependencies. This model can capture the dynamic evolution characteristics of the state parameters of the battery cells at different time points and spatial positions, and realize the comprehensive description and prediction of the state of the target battery.

[0037] S14. According to the measured value of the real-time state of charge of the target battery cell in the target battery, perform a hidden state recursive estimation on the global integration model to determine the estimated value of the state of charge of the target battery cell.

[0038] Among them, the measured value is the state of charge of the target battery cell measured in real time by the direct measurement method (such as the ampere-hour integration method and the power integration method), which is used to assist in estimating the state of charge of the target battery cell, and then obtain the estimated value of the state of charge of the target battery cell.

[0039] Specifically, based on the measured value of the real-time state of charge of the target battery cell in the target battery, the way to perform recursive estimation of the hidden state of the global integration model to determine the estimated value of the state of charge of the target battery cell can be to, based on the measured value of the real-time state of charge of the target battery cell, combine the evolution law of the battery state described by the global integration model in the time and space dimensions, and through an algorithm or model with inference and estimation capabilities, recursively estimate the state of charge of the target battery cell at the current moment, so as to obtain the estimated value.

[0040] S15. Determine the state of charge of the target battery cell according to the historical state parameters and the estimated value.

[0041] Specifically, the way to determine the state of charge of the target battery cell according to the historical state parameters and the estimated value can be to first calculate the spatial correlation parameter between the target battery cell and other battery cells based on the historical state parameters using a spatial correlation algorithm; then, combine the spatial correlation parameter with the estimated value to determine the state of charge of the target battery cell. Among them, the spatial correlation algorithm can be the spatial kernel function in S12, or other spatial correlation algorithms, or a pre-trained model can be used as the spatial correlation algorithm to calculate the spatial correlation parameter.

[0042] In the above solution, since the time kernel function and the spatial kernel function can be established according to the historical state parameters of the battery cell, and the global integration model can be constructed according to the time kernel function and the spatial kernel function, and finally the state of charge of the battery cell can be determined according to the measured value of the real-time state of charge of the target battery cell and the global integration model. In this way, the time kernel function and the spatial kernel function comprehensively characterize the correlation of the battery cell in time evolution and spatial distribution, and then a global integration model with global coupling characteristics is constructed, thereby improving the accuracy of the state of charge determined according to the global integration model. In addition, it is also possible to robustly recursively estimate the state of charge of the battery cell in the presence of observation noise and initial uncertainty, further improving the accuracy and adaptability of the determined state of charge.

[0043] In some embodiments, as Figure 3 shown, the electrical parameter includes the historical state of charge; the way to construct the time kernel function of the target battery according to the sampling time and the electrical parameter can include the following steps: S1201. Determine the standard deviation of the historical state of charge of multiple battery cells.

[0044] Specifically, the way to determine the standard deviation of the historical state of charge of multiple battery cells can be to first determine the average value of the historical state of charge of multiple battery cells; then, determine the standard deviation of the historical state of charge of multiple battery cells according to the historical state of charge of multiple battery cells, the average value, and the total number of data of the historical state of charge.

[0045] For example, it can be determined according to the formula the standard deviation of the historical state of charge of multiple battery cells. Among them, is used to represent the standard deviation of the historical state of charge of multiple battery cells, is used to represent the historical state of charge of any one battery cell, is used to represent the average value of the historical state of charge of multiple battery cells, n is used to represent the total number of data of the historical state of charge.

[0046] S1202. Fit the electrical parameters according to the sampling time to obtain the time-correlation decay rate of the electrical parameters.

[0047] Specifically, the method of fitting the electrical parameters according to the sampling time to obtain the time-correlation decay rate of the electrical parameters can be to construct a function model of the electrical parameters changing with time based on the electrical parameters obtained at multiple sampling times. For example, methods such as exponential decay function, power function or Gaussian process are used to fit it. By analyzing the time factor in the fitting model, the correlation decay characteristics of the electrical parameters in the time dimension can be extracted, and then its time-correlation decay rate can be obtained.

[0048] It should be noted that the time-correlation decay rate is a value that changes with time according to the state of charge during use, and it has the following relationship with the state of charge: when the state of charge changes rapidly, the weight of the historical state parameters is low, and the time-correlation decay rate will become high; when the state of charge changes slowly, the weight of the historical state parameters is high, and the value of the time-correlation decay rate will decrease.

[0049] S1203. Construct the time kernel function of the target battery according to the standard deviation, the time-correlation decay rate and the time difference.

[0050] Among them, the time difference is the time difference between any two sampling times.

[0051] Specifically, the method of constructing the time kernel function of the target battery according to the standard deviation, the time-correlation decay rate and the time difference can be that the output of the time kernel function is determined to be the product of the standard deviation and the first function, where the first function is an exponential function with the base of the natural logarithm and the exponent of the product of the negative time-correlation decay rate and the absolute value of the time difference.

[0052] In some embodiments, the time kernel function can be expressed as: .

[0053] Among them, is used to represent the output of the time kernel function, is used to represent the standard deviation, For representing the decay rate of time correlation, For representing the time difference, is the base of the natural logarithm.

[0054] In the above solution, the standard deviation reflects the consistency of the state of charge among battery cells, the decay rate of time correlation reveals the decay trend of electrical parameters over time, and the time difference quantifies the time interval between samples. The combined effect of these three factors constructs a time kernel function that can effectively capture the state correlation of the target battery at different time points, thereby improving the accuracy and robustness of the subsequent global integration model constructed based on the time kernel function.

[0055] In some embodiments, as Figure 4 shown, the electrical parameters at least include the historical state of charge; the method of constructing the spatial kernel function of the target battery based on the spatial position and electrical parameters may include the following steps: S1211. Determine the standard deviation of the historical state of charge of multiple battery cells.

[0056] Specifically, the method of determining the standard deviation is the same as that in S1201 and will not be elaborated here.

[0057] S1212. Construct the spatial kernel function of the target battery based on the standard deviation, spatial distance, and electrical parameters.

[0058] Among them, the spatial distance is the distance between the spatial positions of any two battery cells among multiple battery cells.

[0059] Specifically, the spatial kernel function can be: .

[0060] Among them, and both represent electrical parameters, and and correspond to different battery cells among multiple battery cells; represents the output of the spatial kernel function, is and the spatial distance between the two corresponding battery cells, represents the standard deviation, is the base of the natural logarithm.

[0061] In the above solution, the standard deviation of the historical state of charge reflects the consistency of the performance of battery cells, and the spatial distance reflects the actual physical layout of battery cells. Introducing both of them and the electrical parameters into the spatial kernel function helps to capture the correlation and coupling effect among battery cells in a local area, thereby improving the accuracy and robustness of the subsequent global integration model constructed based on the spatial kernel function.

[0062] In some embodiments, such as Figure 5 shown, the method of constructing a global integration model according to a temporal kernel function and a spatial kernel function may include the following steps: S131. Perform a rational spectral decomposition on the temporal kernel function to obtain the hidden state variables of the battery cells in the target battery.

[0063] Among them, rational spectral decomposition is a matrix decomposition method that decomposes a matrix into multiple parts, with each part corresponding to a rational irreducible polynomial in the minimal polynomial of the matrix. Through rational spectral decomposition, the matrix is represented as the sum of several projection matrices, and each projection matrix reflects the behavior of the matrix under the corresponding rational irreducible polynomial.

[0064] Specifically, the method of performing a rational spectral decomposition on the temporal kernel function to obtain the hidden state variables of the battery cells in the target battery may be to perform a rational spectral decomposition on the temporal kernel function to obtain the hidden state variables of multiple battery cells in the target battery, where one battery cell corresponds to one hidden state variable, and one hidden state variable corresponds to one rational irreducible polynomial.

[0065] S132. Construct a continuous-time state space model of the target battery according to the hidden state variables, and construct a global state space model according to the continuous-time state space model and the spatial kernel function.

[0066] First, construct a continuous-time state space model of the target battery according to the hidden state variables.

[0067] Specifically, the continuous-time state space model can be expressed as: ; .

[0068] Among them, is used to represent the change rate of the hidden state of the i th battery cell, is used to represent the decay rate of the time correlation of the electrical parameters, is used to represent the hidden state variable of the i th battery cell, is used to represent the Gaussian white noise that drives the state equation, is used to represent the state of charge component including time dynamics of the i th battery cell, is used to represent the standard deviation of the historical states of charge of multiple battery cells.

[0069] Secondly, construct a global state space model according to the continuous-time state space model and the spatial kernel function.

[0070] Specifically, the global state space model can be expressed as: ; .

[0071] Among them, is used to represent the set of discrete state vectors corresponding to the k-th moment after discretization processing, is used to represent k the discrete observation vector at the moment, is used to represent the time interval of discretization processing ; is used to represent the process noise, is used to represent the measurement noise, is used to represent the lower triangular matrix after the Cholesky decomposition of the spatial kernel matrix (calculated according to the spatial kernel function), is used to represent the Kronecker product, is the base of the natural logarithm, is used to represent the decay rate of the time correlation of electrical parameters, is used to represent the standard deviation of the historical state of charge of multiple battery cells.

[0072] S133. Integrate the battery cells in the target battery according to the continuous-time state space model and the global state space model to obtain a global integration model.

[0073] In some embodiments, the global integration model can be:[[]] ; .

[0074] Among them, is used to represent the decay rate of the time correlation of electrical parameters, is used to represent the standard deviation of the historical state of charge of multiple battery cells, is used to represent the time interval of discretization processing ; is used to represent the process noise, is used to represent the measurement noise, is used to represent the lower triangular matrix after the Cholesky decomposition of the spatial kernel matrix, is used to represent the Kronecker product, is used to represent the set of discrete state vectors at the moment, is used to represent k the discrete observation vector at the moment, is the base of the natural logarithm; is M an n-dimensional global identity matrix, M is the number of battery cells in the target battery.

[0075] In the above solution, by rationally decomposing the time kernel function to extract the hidden state variables of the battery cell, the deep analysis of the battery's dynamic characteristics is realized; the continuous-time state space model constructed using the hidden state variables can accurately depict the behavior pattern of the battery cell evolving over time; further combining the spatial kernel function to construct a global state space model effectively integrates time and space information and enhances the model's ability to describe the overall behavior of the system. Then, by globally integrating the model to uniformly represent the co-evolution process of each cell in the battery system, the accuracy of the state of charge estimation is improved.

[0076] In some embodiments, as Figure 6 shown, according to the measured value of the real-time state of charge of the target battery cell in the target battery, the method for performing a hidden state recursive estimation on the global integration model to determine the estimated value of the state of charge of the target battery cell may include the following steps: S141. Use the Kalman filter to perform a hidden state recursive estimation on the global integration model according to the measured value of the real-time state of charge of the target battery cell in the target battery, and obtain the optimal posterior estimation value at the current moment.

[0077] Among them, the Kalman filter recursive update process is as follows: Prediction step: .

[0078] Update step: .

[0079] Among them, A is the discrete state transition matrix, ; is the measured value of the real-time state of charge of the target battery cell in the target battery; is the Kalman gain, which is used to dynamically balance the confidence levels of the estimated value and the measured value, ; C is the output matrix of the global integration model, , is C the transpose matrix of; R is the observation noise covariance matrix, ; is the measurement noise variance; is the output of the update step, is the output of the update step.

[0080] Specifically, the method of using Kalman filtering to perform recursive estimation of the hidden state of the global integration model based on the measured value of the real-time state of charge of the target battery cell in the target battery to obtain the optimal posterior estimation value at the current moment can be as follows: According to the prediction step of Kalman filtering, the state transition matrix is used to perform time update on the hidden state to obtain the prior estimation and covariance matrix at the current moment. Subsequently, in combination with the measured value of the state of charge of the target battery cell at the current moment, the Kalman gain is calculated in the update step, and the prior estimation is weighted and corrected using this gain to obtain the optimal posterior estimation value at the current moment. This recursive process is cycled at each sampling moment to achieve real-time dynamic estimation of the state of charge of the target battery. S142. Obtain the output matrix of the global integration model.

[0081] Specifically, the output matrix of the global integration model is the one in S141 .

[0082] S143. Determine the estimated value as the product of the output matrix and the optimal posterior estimation value.

[0083] Specifically, the calculation method of the estimated value can be . Among them, is the estimated value, C is the output matrix, is the optimal posterior estimation value at the current moment.

[0084] In the above solution, by introducing Kalman filtering to perform recursive estimation of the hidden state of the global integration model, the real-time measurement value of the target battery cell and the model prediction information are effectively fused, thereby improving the real-time performance and accuracy of state estimation. Kalman filtering can obtain the optimal posterior estimation value at the current moment under the conditions of noise and uncertainty, ensuring the high reliability of the dynamic update of the hidden state.

[0085] In some embodiments, as Figure 7 shown, the method of determining the state of charge of the target battery cell according to the historical state parameters and the estimated value may include the following steps: S151. According to the historical state parameters, determine the spatial correlation vector between the target battery cell and multiple battery cells, and the spatial kernel matrix between other battery cells.

[0086] Among them, the other battery cells are the battery cells in the target battery except the target battery cell.

[0087] Specifically, the method for determining the spatial kernel matrix between other battery cells based on historical state parameters may be to input the historical state parameters of other battery cells into a spatial kernel function to obtain the spatial kernel matrix. The method for determining the spatial correlation vector between the target battery cell and multiple battery cells based on historical state parameters may be to first determine a global spatial kernel matrix according to the historical state parameters and the spatial kernel function of all battery cells in the target battery, and then extract the spatial correlation vector related to the target battery cell from the global spatial kernel matrix.

[0088] S152. Determine the state of charge of the target battery cell as the product of the estimated value, the spatial correlation vector, and the inverse matrix of the spatial kernel matrix.

[0089] Specifically, the state of charge of the target battery cell can be expressed as: . Wherein, is used to represent the state of charge of the target battery cell, is used to represent the spatial correlation vector, , is used to represent the set of electrical parameters of the battery cell; is used to represent the electrical parameter of the target battery cell; is used to represent the inverse matrix of the spatial kernel matrix, is used to represent the estimated value.

[0090] In the above solution, the spatial correlation vector reflects the degree of association between the target battery cell and other cells in historical behavior, and the spatial kernel matrix depicts the covariance relationship of the overall spatial structure. Fusing these spatial information with the estimated value helps to achieve a robust inference of the state of the target battery cell in the case of partial data loss or inaccurate measurement, thereby enhancing the reliability and robustness of the state of charge estimation in actual operating scenarios.

[0091] In some embodiments, before constructing the temporal kernel function and spatial kernel function of the target battery based on historical state parameters, the method further includes: detecting abnormal parameters in the historical state parameters and removing the abnormal parameters. Specifically, the method for detecting abnormal parameters in the historical state parameters may be to detect abnormal parameters in the historical state parameters by setting a reasonable threshold range, or to identify outliers that significantly deviate from the normal distribution using statistical methods (such as mean and standard deviation, box plot analysis), or to detect abnormal parameters in the historical state parameters using anomaly detection algorithms in machine learning (such as Isolation Forest, Local Outlier Factor). After detecting the abnormal parameters in the historical state parameters, delete them to obtain the historical state parameters to ensure the reliability of the historical state parameters.

[0092] In this way, before constructing the time kernel function and space kernel function of the target battery, detecting and removing abnormal parameters in the historical state parameters helps to exclude outliers caused by sensor failures, data acquisition errors, or extreme working conditions, preventing them from interfering with the subsequent model construction, thereby improving the accuracy and stability of kernel function modeling.

[0093] In some embodiments, detecting abnormal parameters in the historical state parameters includes: detecting abnormal parameters in the historical state parameters by using an outlier factor detection method.

[0094] Specifically, the process of detecting abnormal parameters in the historical state parameters by using the outlier factor detection method may include the following steps: (1) According to Calculate the k reachability distance of each data point (corresponding to a set of historical state parameters of a single battery cell) to each point within the k distance neighborhood. Among them, is used to represent the o reachability distance from data point p to data point k , is used to represent the o distance of point k , is used to represent the distance from point o to point p .

[0095] (2) According to Calculate the local k local reachability density of each data point. Among them, is used to represent the local p local reachability density of point k , and the reciprocal of the average p distance of all points within the k distance neighborhood of point p to point k reachability distance, is used to represent the o reachability distance from data point p to data point k , is used to represent the set of all points within the p neighborhood of point k , including points within the k distance.

[0096] (3) According to Calculate the k local outlier factor of each data point. Among them, is used to represent the p local outlier factor of point kLocal outlier factor, To represent a point p No. k The set of all points in the neighborhood, To represent a point p Part of k Locally accessible density, To represent a point o Part of k Locally accessible density.

[0097] (4) According to the k The local outlier factor determines whether the data point is an abnormal point, that is, whether the historical state parameter of the data point is an abnormal parameter.

[0098] Specifically, according to the k The local outlier factor determines whether the data point is an outlier by k When the local outlier factor is greater than the outlier factor threshold, the data point is determined to be an outlier. The outlier factor threshold is a preset value, for example, a default value or a value set according to actual conditions, for example, the outlier factor threshold is 1.

[0099] In some embodiments, based on the first k The local outlier factor determines whether the data point is an outlier by directly calculating the maximum u The local outlier factor of each data point is used to determine the abnormal point to save computing resources.

[0100] In the above scheme, the outlier factor detection method is used to identify abnormal parameters in the historical state parameters, which helps to accurately find data points that deviate from the normal distribution and ensure the reliability of the data used in modeling.

[0101] In view of the large amount of data and nonlinearity during the battery charging and discharging process, the embodiment of the present application uses the LOF method to eliminate abnormal data, eliminate the influence of abnormal data on the regularity of characteristic quantities, and reduce the running time of the subsequent time kernel function, space kernel function, continuous time state space model, global state space model, and global integration model (hereinafter collectively referred to as the model). In addition, a model is established through a spatiotemporal Gaussian process, and the SOC evolution of the battery pack is modeled as a spatiotemporal Gaussian process. Its mean function characterizes the expected decay trend of SOC, and the covariance function captures the spatial coupling and time dynamic characteristics between cells. The infinite-dimensional Gaussian process is converted into a finite-dimensional Kalman state space model, and real-time SOC estimation is achieved by recursive updating, and the computational complexity is reduced. It also supports online increase and decrease of monitoring cells, and maintains prediction consistency through covariance matrix reconstruction and state projection, avoiding recalculation of historical data, improving the estimation accuracy of the state of charge, and the accuracy of the model. Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation method.

[0102] The embodiments of the present application also provide a device for determining the state of charge of a battery, as Figure 8 described, the device includes: An acquisition module 701, configured to acquire historical state parameters of each battery cell of a target battery; the target battery includes a plurality of battery cells; a construction module 702, configured to construct a time kernel function and a space kernel function of the target battery according to the historical state parameters; the time kernel function is used to characterize the time correlation between the historical state parameters of the plurality of battery cells; the space kernel function is used to characterize the space correlation between the plurality of battery cells; the construction module 702 is further configured to construct a global integration model according to the time kernel function and the space kernel function; a processing module 703, configured to perform a hidden state recursive estimation on the global integration model according to a measured value of the real-time state of charge of a target battery cell in the target battery, and determine an estimated value of the state of charge of the target battery cell; a determination module 704, configured to determine the state of charge of the target battery cell according to the historical state parameters and the estimated value.

[0103] In some embodiments, the historical state parameters at least include a sampling time, an electrical parameter, and a spatial position of the battery cell in the target battery; the construction module 702 is specifically configured to: construct a time kernel function of the target battery according to the sampling time and the electrical parameter; construct a space kernel function of the target battery according to the spatial position and the electrical parameter.

[0104] In some embodiments, the electrical parameter includes a historical state of charge; the construction module 702 is specifically configured to: determine the standard deviation of the historical states of charge of the plurality of battery cells; perform a fitting process on the electrical parameter according to the sampling time to obtain a time correlation decay rate of the electrical parameter; construct a time kernel function of the target battery according to the standard deviation, the time correlation decay rate, and the time difference; the time difference is the time difference between any two sampling times.

[0105] In some embodiments, the time kernel function is: The time kernel function is: ; where is used to represent the output of the time kernel function, is used to represent the standard deviation, is used to represent the time correlation decay rate, is used to represent the time difference, is the base of the natural logarithm.

[0106] In some embodiments, the electrical parameter at least includes the historical state of charge; the construction module 702 is specifically configured to: determine the standard deviation of the historical state of charge of a plurality of battery cells; construct a spatial kernel function of the target battery according to the standard deviation, the spatial distance, and the electrical parameter; the spatial distance is the distance between the spatial positions of any two battery cells among the plurality of battery cells.

[0107] In some embodiments, the spatial kernel function is: ; where and both are used to represent the electrical parameter, and and correspond to different battery cells among the plurality of battery cells; is used to represent the output of the spatial kernel function, is and the spatial distance between the two corresponding battery cells, is used to represent the standard deviation, is the and base of the natural logarithm.

[0108] In some embodiments, the construction module 702 is specifically configured to: perform a rational spectral decomposition on the time kernel function to obtain the hidden state variables of the battery cells in the target battery; construct a continuous-time state space model of the target battery according to the hidden state variables, and construct a global state space model according to the continuous-time state space model and the spatial kernel function; integrate the battery cells in the target battery according to the continuous-time state space model and the global state space model to obtain a global integration model.

[0109] In some embodiments, the continuous-time state space model is: ; ; where is used to represent the change rate of the hidden state of the i th battery cell, is used to represent the time-correlation attenuation rate of the electrical parameter, is used to represent the hidden state variable of the i th battery cell, is used to represent the Gaussian white noise driving the state equation, is used to represent the state-of-charge component including time dynamics of the i th battery cell, is used to represent the standard deviation of the historical state of charge of a plurality of battery cells; The global state space model is: ; ; where is used to represent after discretization processingk The set of discrete state vectors corresponding to the moment for representing k the discrete observation vector at the moment for representing the discretization process time interval; for representing process noise for representing measurement noise for representing the lower triangular matrix after Cholesky decomposition of the spatial kernel matrix for representing the Kronecker product is the base of the natural logarithm.

[0110] In some embodiments, the global integration model is: ; ; where for representing the decay rate of the time correlation of electrical parameters for representing the standard deviation of the historical state of charge of multiple battery cells for representing the discretized time interval for representing process noise for representing measurement noise for representing the lower triangular matrix after Cholesky decomposition of the spatial kernel matrix for representing the Kronecker product for representing the set of discrete state vectors at the moment is the base of the natural logarithm; is M an \(n\)-dimensional global identity matrix M where \(n\) is the number of battery cells in the target battery.

[0111] In some embodiments, the processing module 703 is specifically configured to: use Kalman filtering to perform hidden state recursive estimation on the global integration model according to the measured value of the real-time state of charge of the target battery cell in the target battery, so as to obtain the optimal posterior estimation value at the current moment; obtain the output matrix of the global integration model; and determine the product of the output matrix and the optimal posterior estimation value as the estimation value.

[0112] In some embodiments, the determination module 704 is specifically configured to: determine the spatial correlation vector between the target battery cell and multiple battery cells, and the spatial kernel matrix between other battery cells according to the historical state parameters; the other battery cells are the battery cells in the target battery except the target battery cell; and determine the product of the estimation value, the spatial correlation vector, and the inverse matrix of the spatial kernel matrix as the state of charge of the target battery cell.

[0113] In some embodiments, the apparatus for determining the state of charge of a battery further includes a detection module; the detection module is configured to detect abnormal parameters in the historical state parameters and eliminate the abnormal parameters before constructing the time kernel function and the space kernel function of the target battery according to the historical state parameters.

[0114] In some embodiments, the detection module is specifically configured to detect abnormal parameters in the historical state parameters by using an outlier factor detection method.

[0115] In some embodiments, the target battery is a lithium battery.

[0116] For the description of the features in the corresponding embodiments of the apparatus for determining the state of charge of a battery, reference can be made to the relevant descriptions in the corresponding embodiments of the method for determining the state of charge of a battery, which will not be elaborated here one by one.

[0117] An embodiment of the present application further provides an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any of the above embodiments of the method for determining the state of charge of a battery.

[0118] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps in any of the above embodiments of the method for determining the state of charge of a battery when running.

[0119] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: various media such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disc that can store a computer program.

[0120] An embodiment of the present application further provides a computer program product. The above computer program product includes a computer program, and when the computer program is executed by a processor, the steps in any of the above embodiments of the method for determining the state of charge of a battery are implemented.

[0121] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium. The non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above embodiments of the method for determining the state of charge of a battery are implemented.

[0122] Those skilled in the art may further realize that the units and algorithm steps of each example described in connection with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0123] The above has introduced in detail a method, apparatus, device, and readable storage medium for determining the state of charge of a battery provided in this application. Specific examples are used herein to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and modifications can be made to this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A method for determining the state of charge of a battery, characterized in that, Including: Obtaining historical state parameters of each battery cell of a target battery; the target battery includes a plurality of battery cells; Constructing a time kernel function and a space kernel function of the target battery according to the historical state parameters; the time kernel function is used to characterize the time correlation between the historical state parameters of the plurality of battery cells; The space kernel function is used to characterize the space correlation between the plurality of battery cells; Constructing a global integration model according to the time kernel function and the space kernel function; Performing a hidden state recursive estimation on the global integration model according to the measured value of the real-time state of charge of a target battery cell in the target battery to determine an estimated value of the state of charge of the target battery cell; Determining the state of charge of the target battery cell according to the historical state parameters and the estimated value.

2. The determination method according to claim 1, wherein The historical state parameters at least include sampling time, electrical parameters, and the spatial position of the battery cell in the target battery; The constructing the time kernel function and the space kernel function of the target battery according to the historical state parameters includes: Constructing the time kernel function of the target battery according to the sampling time and the electrical parameters; Constructing the space kernel function of the target battery according to the spatial position and the electrical parameters.

3. The determination method according to claim 2, characterized in that, The electrical parameters include historical state of charge; the constructing the time kernel function of the target battery according to the sampling time and the electrical parameters includes: Determining the standard deviation of the historical state of charge of the plurality of battery cells; Performing a fitting process on the electrical parameters according to the sampling time to obtain the time correlation decay rate of the electrical parameters; Constructing the time kernel function of the target battery according to the standard deviation, the time correlation decay rate, and the time difference; the time difference is the time difference between any two sampling times.

4. The determination method according to claim 3, wherein The time kernel function is: ； wherein, is used to represent the output of the time kernel function, is used to represent the standard deviation, is used to represent the time correlation decay rate, is used to represent the time difference, is the base of the natural logarithm.

5. The determination method according to claim 2, wherein The electrical parameters at least include historical state of charge; the constructing the space kernel function of the target battery according to the spatial position and the electrical parameters includes: Determining the standard deviation of the historical state of charge of the plurality of battery cells; Constructing the space kernel function of the target battery according to the standard deviation, the spatial distance, and the electrical parameters; the spatial distance is the distance between the spatial positions of any two battery cells among the plurality of battery cells.

6. The determination method according to claim 5, characterized in that The space kernel function is: ； Among them, and are both used to represent electrical parameters, and and correspond to different battery cells among multiple battery cells; is used to represent the output of the spatial kernel function, is and the spatial distance between the two battery cells corresponding to is used to represent the standard deviation, is and the base of the natural logarithm.

7. The determination method according to claim 1, characterized in that The constructing the global integration model according to the time kernel function and the space kernel function includes: Performing a rational spectral decomposition on the time kernel function to obtain the hidden state variables of the battery cells in the target battery; Constructing a continuous-time state space model of the target battery according to the hidden state variables, and constructing a global state space model according to the continuous-time state space model and the space kernel function; Integrating the battery cells in the target battery according to the continuous-time state space model and the global state space model to obtain a global integration model.

8. The determination method according to claim 7, characterized in that, The continuous-time state space model is: ； ； Among them, used to represent the change rate of the hidden state of the i nth battery cell, used to represent the time-correlated decay rate of the electrical parameter, used to represent the hidden state variable of the i nth battery cell, used to represent the Gaussian white noise for driving the state equation, used to represent the state-of-charge component including time dynamics of the i nth battery cell, used to represent the standard deviation of the historical state-of-charge of multiple battery cells; The global state space model is: ； ； Among them, used to represent after discretization processing k the set of discrete state vectors corresponding to the moment, used to represent k the discrete observation vector at the moment, used to represent the discretization processing of the time interval; used to represent the process noise, used to represent the measurement noise, used to represent the lower triangular matrix after the Cholesky decomposition of the spatial kernel matrix, used to represent the Kronecker product, is the base of the natural logarithm.

9. The determination method according to claim 8, wherein The global integration model is: ； ； Among them, is used to represent the time-correlated decay rate of electrical parameters, is used to represent the standard deviation of the historical state of charge of multiple battery cells, is used to represent the discretized time interval, is used to represent the process noise, is used to represent the measurement noise, is used to represent the lower triangular matrix after Cholesky decomposition of the spatial kernel matrix, is used to represent the Kronecker product, is used to represent the set of discrete state vectors at time is the base of the natural logarithm; is M the global identity matrix of dimension M is the number of battery cells in the target battery.

10. The determination method according to claim 1, wherein Performing a hidden state recursive estimation on the global integration model according to the measured value of the real-time state of charge of the target battery cell in the target battery, and determining an estimated value of the state of charge of the target battery cell, includes: Using Kalman filtering to perform a hidden state recursive estimation on the global integration model according to the measured value of the real-time state of charge of the target battery cell in the target battery, and obtaining an optimal posterior estimation value at the current moment; Obtaining an output matrix of the global integration model; Determining the product of the output matrix and the optimal posterior estimation value as the estimated value.

11. The determination method according to claim 1, wherein Determining the state of charge of the target battery cell according to the historical state parameter and the estimated value, includes: According to the historical state parameter, determining a spatial correlation vector between the target battery cell and the multiple battery cells, and a spatial kernel matrix between other battery cells; the other battery cells are battery cells in the target battery except the target battery cell; Determining the product of the estimated value, the spatial correlation vector, and the inverse matrix of the spatial kernel matrix as the state of charge of the target battery cell.

12. The determination method according to claim 1, wherein Before constructing the temporal kernel function and the spatial kernel function of the target battery according to the historical state parameter, the method further includes: Detecting abnormal parameters in the historical state parameter and removing the abnormal parameters.

13. The determination method according to claim 12, characterized in that, The detecting abnormal parameters in the historical state parameter includes: Detecting abnormal parameters in the historical state parameter by using an outlier factor detection method.

14. The determination method according to claim 1, wherein The target battery is a lithium battery.

15. A device for determining the state of charge of a battery, characterized in that, Includes: An obtaining module, configured to obtain historical state parameters of each battery cell of a target battery; the target battery includes multiple battery cells; A constructing module, configured to construct a temporal kernel function and a spatial kernel function of the target battery according to the historical state parameter; the temporal kernel function is used to characterize the temporal correlation between the historical state parameters of the multiple battery cells; The spatial kernel function is used to characterize the spatial correlation between the multiple battery cells; The constructing module is further configured to construct a global integration model according to the temporal kernel function and the spatial kernel function; A processing module, configured to perform a hidden state recursive estimation on the global integration model according to the measured value of the real-time state of charge of the target battery cell in the target battery, and determine an estimated value of the state of charge of the target battery cell; A determining module, configured to determine the state of charge of the target battery cell according to the historical state parameter and the estimated value.

16. An electronic device, characterized in that, Includes: A memory, configured to store a computer program; A processor, configured to implement the steps of the method for determining the state of charge of a battery according to any one of claims 1 to 14 when executing the computer program.

17. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, wherein the computer program, when executed by a processor, implements the steps of the method for determining the state of charge of a battery according to any one of claims 1 to 14.

Images