Optimization method of battery parameter identification algorithm and battery parameter identification method and device

Abstract

The invention relates to the technical field of battery management, and particularly provides an optimization method of a battery parameter identification algorithm and a battery parameter identification method and device.The method can comprise the steps that preliminary identification parameters of a battery equivalent circuit model are obtained; taking the initial identification parameter as an initial value of a to-be-updated battery parameter identification algorithm; continuously inputting the initial value into the to-be-updated battery parameter identification algorithm, and obtaining a plurality of covariance matrixes after the to-be-updated battery parameter identification algorithm runs; when the fluctuation values of the covariance matrixes meet a preset condition and the convergence of the to-be-updated battery parameter identification algorithm is completed, obtaining a gain matrix and a to-be-identified parameter vector in the to-be-updated battery parameter identification algorithm after the convergence is completed; and obtaining a battery parameter identification algorithm based on the gain matrix and the to-be-identified parameter vector. According to the embodiment of the invention, the battery parameter identification algorithm can be obtained, so that the accuracy of battery parameter identification can be improved subsequently.

Description

Technical Field

[0001] This application relates to the field of battery management technology, and more specifically, to an optimization method for battery parameter identification algorithms, a battery parameter identification method, and an apparatus. Background Technology

[0002] A battery management system (BMS) is the core control unit of a lithium battery (such as a power battery or an energy storage battery). Essentially, it is an electronic system that integrates "monitoring, protection, equalization, estimation, and communication". Its core objectives are to ensure the safe and stable operation of the battery pack, extend its service life, and optimize its charging and discharging performance.

[0003] Currently, in traditional methods for hybrid parameter identification of battery models in battery management systems, battery identification parameters are obtained by randomly setting initial values ​​for an online parameter identification algorithm and solving it. If the initial values ​​are set improperly, the accuracy of the obtained battery identification parameters will be low, which will affect the reliability of the battery management system when using these parameters to estimate battery-related indicators.

[0004] Therefore, how to provide a more accurate method for identifying battery parameters has become an urgent technical problem to be solved. Summary of the Invention

[0005] The purpose of some embodiments of this application is to provide an optimization method for battery parameter identification algorithms, a battery parameter identification method and apparatus. The technical solutions of the embodiments of this application can improve the accuracy of battery parameter identification, thereby improving the accuracy of battery-related index estimation and ensuring the stability and reliability of the battery management system.

[0006] In a first aspect, some embodiments of this application provide an optimization method for a battery parameter identification algorithm, comprising: obtaining preliminary identification parameters of a battery equivalent circuit model; wherein the preliminary identification parameters include multiple resistance values ​​and multiple capacitance values; using the preliminary identification parameters as initial values ​​for a battery parameter identification algorithm to be updated; continuously inputting the initial values ​​into the battery parameter identification algorithm to be updated to obtain multiple covariance matrices after the battery parameter identification algorithm to be updated has run; when the fluctuation values ​​of the multiple covariance matrices satisfy a preset condition and the battery parameter identification algorithm to be updated converges, obtaining the gain matrix and the parameter vector to be identified in the battery parameter identification algorithm to be updated at the time of convergence; and obtaining a battery parameter identification algorithm based on the gain matrix and the parameter vector to be identified.

[0007] Some embodiments of this application obtain battery model parameters by acquiring preliminary parameters that match the battery type to be identified. These parameters are then continuously input into the battery parameter identification algorithm as initial values ​​to obtain multiple covariance matrices. The fluctuation of these covariance matrices is analyzed to determine whether the updated battery parameter identification algorithm has converged. Once convergence is achieved, the battery parameter identification algorithm is obtained. This application improves upon the traditional method of randomly setting initial values ​​by using a preliminary parameter identification method, thereby reducing the problem of poor battery parameter identification accuracy caused by random setting. Simultaneously, by continuously inputting these initial values ​​into the updated battery parameter identification algorithm to obtain multiple covariance matrices, the converged battery parameter identification algorithm is obtained using these covariance matrices as constraints. This method enables rapid convergence of the battery parameter identification algorithm, allowing for accurate battery parameter identification and further improving the accuracy of subsequent battery-related index estimations, ensuring the stability and reliability of the battery management system.

[0008] In some embodiments, the battery parameter identification algorithm is related to the open-circuit voltage, terminal voltage measurements, the parameter vector to be identified, and the regression vector of the battery equivalent circuit model; the open-circuit voltage and the terminal voltage measurements are related to the preliminary identification parameters.

[0009] Some embodiments of this application can achieve accurate and effective identification of battery parameters by associating the battery parameter identification algorithm with the relevant parameters of the battery equivalent circuit model.

[0010] In some embodiments, obtaining preliminary identification parameters of the battery equivalent circuit model includes: constructing an optimization objective based on the predicted and measured terminal voltage values ​​of the battery equivalent circuit model; and using the optimization objective as a condition, solving the battery equivalent circuit model using a swarm intelligence optimization algorithm to obtain the preliminary identification parameters; wherein the swarm intelligence optimization algorithm includes a particle swarm optimization algorithm.

[0011] In some embodiments of this application, after constructing optimization objectives based on predicted and measured terminal voltage values, a swarm intelligence optimization algorithm is used to solve the battery equivalent circuit model using these objectives. This yields preliminary identification parameters, providing reliable data support for subsequent battery parameter identification. Solving the battery equivalent circuit model using a particle swarm optimization algorithm can address the issues of slow convergence and divergence in this stage.

[0012] In some embodiments, constructing an optimization objective based on the predicted and measured terminal voltage values ​​from the battery equivalent circuit model includes using the root mean square error between the predicted and measured terminal voltage values ​​as the optimization objective.

[0013] Some embodiments of this application use the root mean square error between the predicted and measured terminal voltage values ​​as the optimization target, which can effectively obtain battery model parameters.

[0014] In some embodiments, the battery parameter identification algorithm is based on the least squares method.

[0015] Some embodiments of this application construct a battery parameter identification algorithm using the least squares method to achieve online fine tracking and can identify parameters throughout the entire battery life cycle.

[0016] In some embodiments, the battery equivalent circuit model includes open-circuit voltage, battery internal resistance, and a resistor-capacitor circuit.

[0017] Secondly, some embodiments of this application provide a method for battery parameter identification, including: real-time acquisition of battery data of a battery to be identified; wherein the battery data includes voltage data and current data; inputting the battery data into a battery parameter identification algorithm, and outputting target battery parameters of the battery to be identified; wherein the battery parameter identification algorithm is obtained through any of the method embodiments in the first aspect.

[0018] Thirdly, some embodiments of this application provide an optimization apparatus for a battery parameter identification algorithm, comprising: an initial value acquisition module, used to use the preliminary identification parameters as initial values ​​for the battery parameter identification algorithm to be updated; a running module, used to continuously input the initial values ​​into the battery parameter identification algorithm to be updated, and obtain multiple covariance matrices after the battery parameter identification algorithm to be updated has run; and an update module, used to obtain the gain matrix and the parameter vector to be identified in the battery parameter identification algorithm when the fluctuation values ​​of the multiple covariance matrices meet preset conditions and the battery parameter identification algorithm to be updated converges; and to obtain the battery parameter identification algorithm based on the gain matrix and the parameter vector to be identified.

[0019] Fourthly, some embodiments of this application provide a battery parameter identification device, comprising: a data acquisition module for real-time acquisition of battery data of a battery to be identified; wherein the battery data includes voltage data and current data; and an identification module for inputting the battery data into a battery parameter identification algorithm and outputting target battery parameters of the battery to be identified; wherein the battery parameter identification algorithm is obtained through any of the method embodiments in the first aspect.

[0020] Fifthly, some embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the methods described in any of the embodiments of the first or second aspect.

[0021] In a sixth aspect, some embodiments of this application provide an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, can implement the method as described in any embodiment of the first or second aspect.

[0022] In a seventh aspect, some embodiments of this application provide a computer program product comprising a computer program, wherein the computer program, when executed by a processor, can implement the method described in any embodiment of the first or second aspect. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of some embodiments of this application, the accompanying drawings used in some embodiments of this application will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 Flowchart of an optimization method for a battery parameter identification algorithm provided for some embodiments of this application; Figure 2 A schematic diagram of the battery equivalent circuit model structure is provided for some embodiments of this application; Figure 3 One of the flowcharts for a method of battery parameter identification provided in some embodiments of this application; Figure 4 A flowchart of the PSO algorithm provided for some embodiments of this application; Figure 5 A second flowchart illustrating a method for battery parameter identification provided for some embodiments of this application; Figure 6 Block diagram of the optimization device for the battery parameter identification algorithm provided for some embodiments of this application; Figure 7 A schematic diagram of an electronic device provided for some embodiments of this application. Detailed Implementation

[0025] The technical solutions of some embodiments of this application will now be described with reference to the accompanying drawings.

[0026] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0027] In related technologies, traditional FFRLS (Forgetting Factor Recursive Least Squares) is prone to slow parameter convergence and estimation divergence in the initial stage due to improper initial values ​​of the covariance matrix. This leads to large errors in the obtained battery parameters, affecting the subsequent estimation of battery SOX (a set of key parameters describing the core state of the battery, where X represents different state dimensions), and thus impacting the reliability of the Battery Management System (BMS). In practical applications, unstable battery parameters during the BMS startup phase can cause large fluctuations in the initial error of battery SOC (State of Charge) estimation, affecting the control decisions of the BMS. Simply increasing the initial covariance matrix can accelerate convergence, but it introduces the risk of numerical instability. Furthermore, when the BMS is first powered on or restarted after a long period of shutdown, the lack of historical data for reference means that battery parameter identification takes a long time to reach a stable state; and the method of random initial values ​​cannot adapt to different battery states and aging levels, thus failing to achieve accurate battery parameter identification.

[0028] As can be seen from the above-mentioned related technologies, the accuracy of battery parameter identification in the existing technology is low and its adaptability is relatively limited.

[0029] In view of this, some embodiments of this application provide a method for battery parameter identification. This method first obtains preliminary identification parameters from the battery's equivalent circuit model; these preliminary identification parameters are then used as initial values ​​for the updated battery parameter identification algorithm to be calculated, resulting in a battery parameter identification algorithm that meets the requirements; subsequently, the battery parameters of the battery to be identified can be accurately identified using the battery parameter identification algorithm. After obtaining the battery parameter identification algorithm, embodiments of this application can use it for online identification, obtaining highly accurate battery parameters; this can further improve the accuracy of battery SOX estimation and enhance the reliability of the battery management system; moreover, embodiments of this application are adaptable to batteries with different battery states and aging levels, exhibiting wide applicability.

[0030] The battery parameter identification method provided in this application can be applied to parameter identification in UAV battery BMS systems or other new energy BMS systems. For the same type of battery, in the case of initial startup or long-term inactivity, an equivalent circuit model of the battery can be pre-constructed, and then preliminary identification parameters corresponding to the model under the corresponding conditions can be obtained. These preliminary identification parameters are used as the initial values ​​of the online parameter identification algorithm (i.e., the battery parameter identification algorithm to be updated). When the covariance matrix of the online parameter identification algorithm converges and the parameters stabilize, the online parameter identification algorithm at this point takes over the online parameter identification and saves the identification results. The identification results can be used as the initial values ​​(i.e., battery model parameters) for the next system startup state, so as to perform online parameter identification again.

[0031] The following is in conjunction with the appendix Figure 1 The implementation process of the optimization method of the battery parameter identification algorithm provided in some embodiments of this application is illustrated by way of example.

[0032] Please see the appendix Figure 1 , Figure 1 A flowchart of an optimization method for a battery parameter identification algorithm is provided for some embodiments of this application. The optimization method for the battery parameter identification algorithm may include: S110, obtain preliminary identification parameters of the battery equivalent circuit model; wherein, the preliminary identification parameters include multiple resistance values ​​and multiple capacitance values.

[0033] For example, in a specific embodiment of this application, for the same type of battery, a corresponding battery equivalent circuit model can be pre-constructed for offline parameter identification to obtain preliminary identification parameters.

[0034] In some embodiments of this application, S110 may include: constructing an optimization objective based on the predicted and measured terminal voltage values ​​of the battery equivalent circuit model; and solving the battery equivalent circuit model using a swarm intelligence optimization algorithm with the optimization objective as a condition to obtain the preliminary identification parameters. The battery equivalent circuit model includes open-circuit voltage, battery internal resistance, and a resistor-capacitor circuit.

[0035] For example, in another specific embodiment of this application, a second-order RC equivalent circuit model (as a specific example of a battery equivalent circuit model) is used as the battery model. The structure of the second-order RC equivalent circuit model is as follows: Figure 2 As shown. Figure 2 The second-order RC equivalent circuit model in the figure is based on the open-circuit voltage U. ocvThe battery consists of three parts: the battery's internal resistance R0 (i.e., the battery's ohmic internal resistance) and two RC circuits connected in series (i.e., resistor-capacitor circuits). R1 ​​and C1 are the electrochemical polarization resistor and capacitor, respectively; R2 and C2 are the concentration difference polarization resistor and capacitor, respectively; U1 and U2 are the voltages across the two RC circuits, U... t Let I be the predicted terminal voltage, and I be the current in the second-order RC equivalent circuit model. The state-space characteristic equation of this second-order RC equivalent circuit model is:

[0036] Where SOC(t) represents the remaining charge state at time t.

[0037] Discretizing the state-space characteristic equation yields the following discretized equation:

[0038]

[0039] in, , Both are the time constants of the two RC loops, where T is the sampling time interval, k is the sampling time, and Q is the sampling time. n For battery capacity, SOC k This represents the remaining battery level at sampling time k. SOC k-1 This represents the remaining battery level at sampling time k-1. U 1,k , U 2,k The voltages of the two RC circuits at sampling time k. SOC k-1 This represents the remaining battery level at sampling time k-1. U 1,k-1 , U 2,k-1 Let I be the voltage of the two RC circuits at sampling time k-1. k-1 Let I be the current at sampling time k-1. k The current at sampling time k. U t,k , U ocv,k These are the predicted terminal voltage and open-circuit voltage at sampling time k.

[0040] After constructing the second-order RC equivalent circuit model as described above, the initial preliminary identification parameters are obtained based on it.

[0041] For example, in a specific embodiment of this application, for the same type of battery, an optimization target is constructed by combining battery capacity, experimental data obtained from HPPC experiments (i.e., terminal voltage measurements) with terminal voltage predictions. A swarm intelligence optimization algorithm is then used for offline parameter identification to obtain high-precision battery model parameters. The swarm intelligence optimization algorithm can be Particle Swarm Optimization (PSO), Artificial Peak Swarm Optimization (ABC), or Grey Wolf Optimization (GWO), etc.

[0042] The optimization objective is constructed based on the predicted and measured terminal voltage values ​​of the battery equivalent circuit model, including using the root mean square error between the predicted and measured terminal voltage values ​​as the optimization objective.

[0043] For example, in a specific embodiment of this application, in the parameter identification of the second-order RC equivalent circuit model of a lithium-ion battery, the root mean square error (RMSE) between the predicted terminal voltage and the experimentally measured value (i.e., the measured terminal voltage) is used as the optimization objective function to evaluate the fitting accuracy of the parameter set. It is understood that, in addition to using RMSE as the optimization objective function, mean square error or other error parameters can also be used, and the embodiments of this application are not limited to this.

[0044] Among them, the swarm intelligence optimization algorithm is the Particle Swarm Optimization (PSO) algorithm.

[0045] Specifically, the core of the particle swarm optimization algorithm lies in guiding the population towards the optimal region by tracking the historical best solutions of individual particles and the global best solution of the swarm. This algorithm dynamically updates the particle velocity (V) based on an iterative formula. i ) and location (X) i The iterative formula for this algorithm is:

[0046] in, w For inertial weights, c 1 and c 2 is the learning factor. r 1 and r 2 is a random number uniformly distributed within the interval [0,1]. For particles d speed, For particles d Location, P id It is the best position in individual history. P gdIt is a global optimum for the population. The population size is set to 100, and the maximum number of iterations is 500. To reduce the impact of algorithm randomness and avoid local optima, multiple independent repeated runs are performed, and the result with the smallest RMSE is selected as the final identification parameter. Through the above optimization process, five key parameters characterizing the dynamic characteristics of the battery in the second-order RC equivalent circuit model (as a specific example of preliminary identification parameters) R0, R1, R2, C1, and C2 can be accurately obtained.

[0047] S120, the preliminary identification parameters are used as the initial values ​​for the battery parameter identification algorithm to be updated.

[0048] For example, the five key parameters identified by PSO can be used as the initial values ​​for subsequent battery parameter identification algorithms, instead of the traditional random initial value scheme.

[0049] S130, the initial value is continuously input into the battery parameter identification algorithm to be updated, and multiple covariance matrices are obtained after the battery parameter identification algorithm to be updated is run.

[0050] Optionally, in some embodiments of this application, only one initial value is continuously input into the battery parameter identification algorithm to be updated, so as to optimize the battery parameter identification algorithm in the future, thereby achieving rapid convergence of the battery parameter identification algorithm and improving the subsequent battery identification accuracy.

[0051] S140, when the fluctuation values ​​of the plurality of covariance matrices meet the preset conditions and the battery parameter identification algorithm to be updated converges, the gain matrix and the parameter vector to be identified in the battery parameter identification algorithm to be updated at the time of convergence are obtained; the battery parameter identification algorithm is obtained based on the gain matrix and the parameter vector to be identified. The battery parameter identification algorithm is constructed based on the least squares method.

[0052] Specifically, the battery parameter identification algorithm is related to the open-circuit voltage and terminal voltage measurements of the battery equivalent circuit model, the parameter vector to be identified corresponding to the battery parameters, and the regression vector; the open-circuit voltage and the terminal voltage measurements are related to the battery model parameters.

[0053] The implementation process of S120~S140 is illustrated below by example.

[0054] For example, in a specific embodiment of this application, an online parameter identification algorithm is constructed based on the least squares method (FFRLS). The obtained R0, R1, R2, C1, and C2 are used as the initial values ​​of the FFRLS algorithm and continuously input into it. The fluctuation of multiple covariance matrices is used as a constraint to update and solve the parameters within the FFRLS algorithm, resulting in a converged battery parameter identification algorithm.

[0055] As a specific example, the relevant formulas for the FFRLS algorithm are as follows: ;in, , For the regression vector, The parameter vector to be identified.

[0056] The parameter update steps in the above formula include:

[0057] Where K is the gain matrix and P is the error covariance matrix (or simply covariance matrix). Let y be the data vector, y be the output vector of the identification result, I be the identity matrix, and the forgetting factor λ ranges from 0.95 to 1.

[0058] During parameter identification, the initial parameters are identified as the initial values ​​for the FFRLS algorithm. These initial values ​​are continuously input into the FFRLS algorithm to update its internal parameters online, resulting in multiple error covariance matrices. When the error covariance matrix tends to a stable value, which can be understood as the fluctuation value of the error covariance matrix meeting a preset condition, the FFRLS algorithm is considered to have converged. After convergence, the gain matrix K and the initial values ​​are obtained. The FFRLS algorithm, once converged, becomes the battery parameter identification algorithm for subsequent online parameter identification. Simultaneously, usable data is stored for the next system power-on. For example, the adjacent errors (i.e., fluctuation values) of the covariance in the error covariance matrix within 10 sampling points are recorded. If all adjacent errors tend to 0 (as a specific example of a stable value or preset condition), the algorithm is confirmed to have converged, and the battery parameters R0, R1, R2, C1, and C2 are output. It is understood that the stable value and preset condition can be set according to actual needs, and the embodiments of this application are not limited thereto.

[0059] In some embodiments of this application, the battery parameter identification method further includes: real-time acquisition of battery data of the battery to be identified; wherein the battery data includes voltage data and current data; inputting the battery data into a battery parameter identification algorithm, and outputting the target battery parameters of the battery to be identified.

[0060] For example, in practical applications, the converged FFRLS algorithm obtained above can be embedded into the BMS. Based on the battery voltage and current data collected in real time by the microcontroller, the converged FFRLS algorithm can be used to update the five battery parameters R0, R1, R2, C1, and C2 online. These five battery parameters are then imported into the subsequent battery SOC and SOH estimation, providing the BMS system with accurate and stable battery parameters.

[0061] The following is in conjunction with the appendix Figure 3 The implementation process of battery parameter identification provided by some embodiments of this application is illustrated by way of example.

[0062] Please see the appendix Figure 3 , Figure 3 A flowchart illustrating a method for battery parameter identification provided for some embodiments of this application.

[0063] The above process is illustrated below by example.

[0064] S310 is used to perform charge and discharge experiments on the battery equivalent circuit model to obtain the terminal voltage measurement value.

[0065] S320 uses the root mean square error between the predicted and measured terminal voltage values ​​as the optimization objective and employs the PSO algorithm to perform offline parameter identification on the battery equivalent circuit model, obtaining preliminary identification parameters.

[0066] S330, the preliminary identification parameters are used as the initial values ​​of the battery parameter identification algorithm to be updated; and the initial values ​​are continuously input into the battery parameter identification algorithm to be updated to obtain multiple covariance matrices after the battery parameter identification algorithm to be updated runs.

[0067] S340: When the fluctuation values ​​of multiple covariance matrices meet the preset conditions and the battery parameter identification algorithm to be updated converges, obtain the gain matrix and the FFRLS algorithm corresponding to the parameter vector to be identified in the battery parameter identification algorithm to be updated when the convergence is complete.

[0068] The S350 acquires the voltage and current data of the battery to be identified in real time, collected by the microcontroller.

[0069] The S360 inputs voltage and current data into the FFRLS algorithm and outputs the target battery parameters for the battery to be identified.

[0070] S370 uses target battery parameters to estimate the SOC of the BMS.

[0071] The target battery parameters can be used as the initial values ​​for the next online parameter identification.

[0072] It is understood that the specific implementation process of S310~S370 can refer to the method embodiment provided above. To avoid repetition, detailed descriptions are omitted here. Among them, S310~S340 can also be executed independently, and can be directly applied for parameter identification after obtaining the FFRLS algorithm.

[0073] The following is in conjunction with the appendix Figure 4 The specific process of S320 provided by some embodiments of this application is illustrated by way of example.

[0074] Please see the appendix Figure 4 , Figure 4 A flowchart of the PSO algorithm in S320 provided for some embodiments of this application.

[0075] The above process is illustrated below by example.

[0076] S410, initialize the particle swarm position and velocity.

[0077] S420 calculates the fitness of each particle.

[0078] S430, update the individual optimal and global optimal.

[0079] S440 updates the velocity and position of each particle.

[0080] S450: Determine if the termination condition is met. If yes, proceed to S460; otherwise, return to S420.

[0081] The termination condition is whether the RMSE of the above-mentioned optimization objective function is less than a preset value.

[0082] S460 outputs the global optimal solution, i.e., the battery model parameters.

[0083] Among them, the battery model parameters are preliminary identification parameters.

[0084] The following is in conjunction with the appendix Figure 5 The present application provides an exemplary description of the specific process for battery parameter identification in some embodiments.

[0085] Please see the appendix Figure 5 , Figure 5 A flowchart illustrating a method for battery parameter identification provided for some embodiments of this application.

[0086] Step 1: First, perform a battery charge-discharge experiment on the equivalent circuit model of the drone battery to obtain experimental voltage values ​​for subsequent offline parameter identification. Step 2: After offline parameter identification using PSO, obtain the five initial parameters (or initial values, or identification results) from Step 3, namely R0, R1, R2, C1, and C2. When online parameter identification is required, execute Step 4 to initialize the system and load the PSO identification results. Simultaneously, execute Step 5 to have the microcontroller collect battery data from the equivalent circuit model of the drone battery, such as battery voltage and current. Step 6: Based on the PSO identification results and battery data, construct the data vector required by the FFRLS algorithm, initialize the error covariance matrix, initialize the parameter vector to be identified, update the gain matrix, update the parameter estimates, and update the error covariance matrix. Step 7: Determine whether the error covariance matrix has converged to a stable value. If so, execute Step 8 to output the battery parameters and obtain the FFRLS algorithm; otherwise, return to Step 6 to continue the related operations in a loop. The battery parameters will also be stored to return to the initial values ​​for the next online identification of the drone battery parameters in step six.

[0087] As can be seen from the above embodiments of this application, this application avoids the slow convergence speed and divergence problems of FFRLS in the initial stage by initializing parameters offline through PSO, enabling parameter identification to quickly enter a stable state and improving the convergence and stability of the algorithm. This is particularly important for the mission characteristics of UAVs with frequent start-stop operations, improving the system response speed. PSO provides a high-precision initial starting point, while FFRLS is responsible for online fine tracking, which can identify parameters throughout the entire battery life cycle, fully describing the state of the battery at different times, and realizing high-precision parameter tracking throughout the battery life cycle; moreover, offline parameters provide better initial values, and online identification is optimized on this basis, reducing initial errors and improving the accuracy of parameter identification, providing high-precision parameters for the BMS system, and improving the overall estimation capability.

[0088] Please refer to Figure 6 , Figure 6 This document illustrates a block diagram of an optimization apparatus for a battery parameter identification algorithm provided in some embodiments of this application. It should be understood that this optimization apparatus corresponds to the method embodiments described above and is capable of executing the various steps involved in the method embodiments. The specific functions of this optimization apparatus can be found in the description above; detailed descriptions are omitted here to avoid repetition.

[0089] Figure 6The optimization device for the battery parameter identification algorithm includes at least one software function module that can be stored in a memory or embedded in the optimization device in the form of software or firmware. The optimization device includes: an initial parameter acquisition module 610, used to acquire preliminary identification parameters of the battery equivalent circuit model; wherein the preliminary identification parameters include multiple resistance values ​​and multiple capacitance values; an initial value acquisition module 620, used to use the preliminary identification parameters as initial values ​​for the battery parameter identification algorithm to be updated; a running module 630, used to continuously input the initial values ​​into the battery parameter identification algorithm to be updated, and acquire multiple covariance matrices after the algorithm has run; and an update module 640, used to acquire the gain matrix and the parameter vector to be identified in the algorithm when the fluctuation values ​​of the multiple covariance matrices meet preset conditions and the algorithm converges; and to obtain the battery parameter identification algorithm based on the gain matrix and the parameter vector to be identified.

[0090] Some embodiments of this application also provide a battery parameter identification device, including: a data acquisition module for real-time acquisition of battery data of a battery to be identified; wherein the battery data includes voltage data and current data; and an identification module for inputting the battery data into a battery parameter identification algorithm and outputting target battery parameters of the battery to be identified; wherein the battery parameter identification algorithm is obtained through any of the above method embodiments.

[0091] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the aforementioned method, and will not be elaborated further here.

[0092] Some embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can perform the operation of any of the methods corresponding to the methods provided in the above embodiments.

[0093] Some embodiments of this application also provide a computer program product, which includes a computer program, wherein when the computer program is executed by a processor, it can implement the operation of any of the methods corresponding to the above embodiments provided in the above embodiments.

[0094] like Figure 7As shown, some embodiments of this application provide an electronic device 700, which includes a memory 710, a processor 720, and a computer program stored in the memory 710 and executable on the processor 720. When the processor 720 reads the program from the memory 710 via a bus 730 and executes the program, it can implement the methods of any of the above embodiments.

[0095] Processor 720 can process digital signals and can include various computing architectures. For example, it can be a complex instruction set computer architecture, a reduced instruction set computer architecture, or an architecture that implements multiple instruction set combinations. In some examples, processor 720 can be a microprocessor.

[0096] The memory 710 can be used to store instructions executed by the processor 720 or data related to the execution of instructions. These instructions and / or data may include code for implementing some or all of the functions of one or more modules described in the embodiments of this application. The processor 720 of this disclosure embodiment can be used to execute the instructions in the memory 710 to implement the methods shown above. The memory 710 includes dynamic random access memory, static random access memory, flash memory, optical memory, or other memories well known to those skilled in the art.

[0097] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0098] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0099] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. An optimization method for a battery parameter identification algorithm, characterized in that, include: Obtain preliminary identification parameters for the battery equivalent circuit model; wherein, the preliminary identification parameters include multiple resistance values ​​and multiple capacitance values; The preliminary identification parameters are used as the initial values ​​for the battery parameter identification algorithm to be updated; The initial value is continuously input into the battery parameter identification algorithm to be updated, and multiple covariance matrices are obtained after the battery parameter identification algorithm to be updated is run. When the fluctuation values ​​of the plurality of covariance matrices meet the preset conditions and the battery parameter identification algorithm to be updated converges, the gain matrix and the parameter vector to be identified in the battery parameter identification algorithm to be updated at the time of convergence are obtained. A battery parameter identification algorithm is obtained based on the gain matrix and the parameter vector to be identified; the battery parameter identification algorithm is constructed based on the least squares method.

2. The method as described in claim 1, characterized in that, The battery parameter identification algorithm is related to the open-circuit voltage, terminal voltage measurement values, the parameter vector to be identified, and the regression vector of the battery equivalent circuit model; the open-circuit voltage and the terminal voltage measurement values ​​are related to the preliminary identification parameters.

3. The method according to any one of claims 1-2, characterized in that, The preliminary identification parameters for obtaining the battery equivalent circuit model include: An optimization objective is constructed based on the predicted and measured terminal voltage values ​​of the battery equivalent circuit model. Using the optimization objective as a condition, the battery equivalent circuit model is solved using a swarm intelligence optimization algorithm to obtain the preliminary identification parameters; wherein, the swarm intelligence optimization algorithm includes the particle swarm optimization algorithm.

4. The method as described in claim 3, characterized in that, The optimization objective is constructed based on the predicted and measured terminal voltage values ​​from the battery equivalent circuit model, including: The root mean square error between the predicted terminal voltage and the measured terminal voltage is used as the optimization objective.

5. The method as described in claim 1 or 2, characterized in that, The battery equivalent circuit model includes open-circuit voltage, battery internal resistance, and resistor-capacitor circuit.

6. A method for identifying battery parameters, characterized in that, include: Real-time acquisition of battery data of the battery to be identified; wherein, the battery data includes voltage data and current data; The battery data is input into the battery parameter identification algorithm, and the target battery parameters of the battery to be identified are output; wherein, the battery parameter identification algorithm is obtained by the method of any one of claims 1-5.

7. An optimization device for a battery parameter identification algorithm, characterized in that, include: An initial parameter acquisition module is used to acquire preliminary identification parameters of the battery equivalent circuit model; wherein, the preliminary identification parameters include multiple resistance values ​​and multiple capacitance values; The initial value acquisition module is used to use the preliminary identification parameters as the initial values ​​of the battery parameter identification algorithm to be updated; The running module is used to continuously input the initial value into the battery parameter identification algorithm to be updated, and to obtain multiple covariance matrices after the battery parameter identification algorithm to be updated has been run. The update module is used to obtain the gain matrix and the parameter vector to be identified in the battery parameter identification algorithm when the fluctuation values ​​of the multiple covariance matrices meet the preset conditions and the battery parameter identification algorithm to be updated converges; and to obtain the battery parameter identification algorithm based on the gain matrix and the parameter vector to be identified; the battery parameter identification algorithm is constructed based on the least squares method.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is executed by a processor to perform the method as described in any one of claims 1-6.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored on the memory and running on the processor, wherein the computer program is executed by the processor to perform the method as described in any one of claims 1-6.

Images