Method and device for determining optimal parameter value and nonvolatile storage medium

Abstract

The invention discloses a method and device for determining an optimal parameter value and a nonvolatile storage medium. The method comprises the steps that an equivalent circuit model of a to-be-tested battery is acquired, and the equivalent circuit model is used for indicating the change state of electrical parameters of the to-be-tested battery in the charging and discharging process; according to the electrical parameters, determining a state space model of the to-be-tested battery, the state space model being used for indicating an incidence relation among the input current of the to-be-tested battery, the output voltage of the to-be-tested battery and the electrical parameters; the state space model is analyzed through a particle swarm optimization algorithm, an optimal parameter value is obtained, a complex order operator is fused in the particle swarm optimization algorithm, and the optimal parameter value is the numerical value of the electrical parameter enabling the error value to be minimum. According to the method and the device, the technical problem that the parameters of the battery model cannot be accurately identified due to the problems of low precision, poor robustness and the like existing in the identification method of the parameters of the battery model in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular, to a method and device for determining an optimal parameter value, and a non-volatile storage medium. Background Art

[0002] High-precision parameter identification of lithium battery models is the basis and guarantee for accurate estimation of various states of power batteries. By identifying the model parameters of lithium batteries, potential battery safety hazards such as overheating, fire, explosion, etc. can be discovered and resolved in a timely manner to ensure the safety of drivers and passengers. An accurate battery model can also avoid battery overcharging and over-discharging problems caused by battery overload, thereby extending the battery life and improving the overall performance of electric vehicles. At present, the identification of battery equivalent circuit model parameters based on particle swarm optimization algorithm has become a research hotspot, but the methods for identifying battery model parameters in related technologies still have problems such as low accuracy, weak adaptability and poor robustness.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a method and device for determining optimal parameter values, and a non-volatile storage medium, so as to at least solve the technical problem that the battery model parameter identification method in the related art cannot accurately identify the battery model parameters due to the low precision and poor robustness.

[0005] According to one aspect of an embodiment of the present application, a method for determining an optimal parameter value is provided, comprising: obtaining an equivalent circuit model of a battery to be tested, wherein the equivalent circuit model is used to indicate the changing state of electrical parameters of the battery to be tested during the charging and discharging process, and during the charging and discharging process, the state of the battery to be tested includes: an electrochemical polarization state and a concentration difference polarization state, and the electrical parameters include: an inherent resistance of the battery to be tested, an open circuit voltage of the battery to be tested, a first resistance and a first capacitance of the battery to be tested in the electrochemical polarization state, and a second resistance and a second capacitance of the battery to be tested in the concentration difference polarization state; determining a state space model of the battery to be tested according to the electrical parameters, wherein the state space model is used to indicate the correlation between the input current of the battery to be tested, the output voltage of the battery to be tested, and the electrical parameters; analyzing the state space model using a particle swarm optimization algorithm to obtain an optimal parameter value, wherein a complex-order operator is integrated into the particle swarm optimization algorithm, and the optimal parameter value is the value of the electrical parameter that minimizes the error value, and the error value is the error value between the current measured voltage value and the current predicted voltage value of the battery to be tested at the current moment, and the current predicted voltage value is a voltage value determined according to the equivalent circuit model and the optimal parameter value.

[0006] Optionally, determining the state space model of the battery to be tested according to the electrical parameters includes: determining the current value of the battery to be tested at the current moment; determining the first voltage according to the first resistance, the first capacitance and the current value, and determining the second voltage according to the second resistance and the second capacitance, wherein the first voltage is used to indicate the electrochemical polarization effect of the battery to be tested during the charging and discharging process, and the second voltage is used to indicate the concentration range effect of the battery to be tested during the charging and discharging process; obtaining the state of charge SOC of the battery to be tested at the current moment, and determining the state variables of the state space model according to the SOC, the first voltage and the second voltage; determining an objective function for indicating the correlation between the SOC and the open circuit voltage of the battery to be tested, and determining the output variables of the state space model according to the objective function, the first voltage, the second voltage, the current value and the inherent resistance; determining the current value as the input variable of the state space model, and determining the state space model according to the state variables, the output variables and the input variables.

[0007] Optionally, a particle swarm optimization algorithm is used to analyze the state space model to obtain the optimal parameter value, including: obtaining multiple groups of historical state information of the state space model at multiple historical moments before the current moment, and generating multiple first target particle swarms to be optimized based on the multiple groups of historical state information, wherein each first target particle swarm to be optimized includes multiple particles to be optimized, each group of historical state information is a particle to be optimized, each historical moment corresponds to a group of historical state information, each group of historical state information includes at least: the historical position of the particle to be optimized at multiple historical moments before the current moment, the historical position is used to indicate the value of the particle to be optimized at the historical moment, each particle to be optimized includes: the value of the first type of parameter included in the state variable in the state space model and the value of the second type of parameter included in the output variable in the state space model, wherein the first type of parameters include: a first capacitor, a first resistor, a second capacitor and the second resistor, the values ​​of the first type of parameters include: the capacitance value of the first capacitor, the first fractional order of the first capacitor, the resistance value of the first resistor, the capacitance value of the second capacitor, the second fractional order of the second capacitor and the resistance value of the second resistor, the first fractional order is used to indicate the frequency response characteristics of the first capacitor, and the second fractional order is used to indicate the frequency response characteristics of the second capacitor, the second type of parameters include: multiple coefficients in the objective function of generating output variables, and the values ​​of the second type of parameters include: multiple values ​​corresponding to the multiple coefficients; using a particle swarm optimization algorithm to iteratively update multiple first target particle swarms to be optimized until the number of iterations reaches a preset number of iterations, and output a target particle swarm, which is the result of the last iterative update; determining the error value corresponding to each particle in the target particle swarm, determining the particle corresponding to the smallest error value as the target particle, and determining the multiple values ​​contained in the target particle as the optimal parameter values.

[0008] Optionally, a particle swarm optimization algorithm is used to iteratively update multiple first target particle swarms to be optimized, including: for each first target particle swarm to be optimized, determining an individual optimal position and a group optimal position according to historical state information, the individual optimal position is the historical position of each particle to be optimized when the historical error value is the smallest, the historical error value is the root mean square difference between multiple predicted voltage values ​​and multiple measured voltage values ​​corresponding to multiple historical moments, and the group optimal position is determined from multiple individual optimal positions; updating the speed and position of each particle to be optimized according to the individual optimal position and the group optimal position to obtain a speed update result and a position update result, wherein the speed update result is an adjustment direction and an adjustment value of the particle to be optimized when performing iterative update, and the adjustment direction includes: increase and decrease, position The update result is set to be the value of the particles to be optimized after iterative update; the update method of the first target particle group to be optimized is determined according to the speed update result and the position update result, and the first target particle group to be optimized is updated according to the update method, and the update method includes: a first update method and a second update method, the first update method is used to indicate to increase the number of particles to be optimized contained in the first target particle group to be optimized, and the second update method is used to indicate to reduce the number of particles to be optimized contained in the first target particle group to be optimized; determine the particle count of each first target particle group to be optimized, and determine the second target particle group to be optimized for the next iterative update according to the particle count, wherein the particle count is the number of particles to be optimized contained in the first target particle group to be optimized after being updated according to the update method.

[0009] Optionally, updating the speed and position of each particle to be optimized according to the individual optimal position and the group optimal position includes: updating the speed of each particle to be optimized according to the following formula: in, represents the complex order operator, α1 represents the first fractional order, β1 represents the second fractional order, represents the speed update result of particle i to be optimized after the kth iteration update, k represents the number of iterations, c1 and c2 are different preset values, r1 and r2 are random numbers within the preset value range, represents the individual best position of particle i to be optimized during the k-th iteration update, represents the position of particle i to be optimized before the kth iteration update, Represents the optimal position of the group during the k-th iteration update.

[0010] Optionally, updating the speed and position of each particle to be optimized according to the individual optimal position and the group optimal position also includes: updating the position of each particle to be optimized according to the following formula: Where k represents the number of iterations, represents the position update result of particle i to be optimized after the kth iteration update, represents the position of particle i to be optimized during the k-th iteration update, Represents the current velocity of particle i to be optimized during the k-th iterative update.

[0011] Optionally, determining an update method of the first target particle swarm to be optimized according to the speed update result and the position update result includes: determining an error value corresponding to each particle to be optimized after this iterative update according to the following formula: Wherein, E represents the error value, k represents the number of iteration updates, M represents the total number of iteration updates, V0(k) is the measured voltage value of the battery under test when performing the kth iteration update, is a voltage value determined according to the equivalent circuit model and the speed update result and the position update result obtained by the k-th iterative update of the particle i to be optimized; the error value is compared with the previous error value to obtain a comparison result, and an update method is determined according to the comparison result, wherein the previous error value is an error value determined according to the previous iterative update result of the particle to be optimized; when the comparison result indicates that the error value is less than the previous error value, the update method is determined to be the first update method; when the comparison result indicates that the error value is greater than or equal to the previous error value, the second update method is determined.

[0012] Optionally, updating the first target particle group to be optimized according to the updating method includes: updating the first target particle group to be optimized according to the first updating method, and updating the first target particle group to be optimized according to the second updating method, wherein updating the first target particle group to be optimized according to the first updating method includes: determining a data set consisting of a velocity update result and a position update result as a new particle; adding the new particle to the particle group to be optimized, and updating the particle count; updating the first target particle group to be optimized according to the second updating method includes: determining an error value corresponding to each particle to be optimized among a plurality of particles to be optimized currently contained in the first target particle group to be optimized; determining the particle to be optimized corresponding to the largest error value as the worst particle; deleting the worst particle, and updating the particle count.

[0013] Optionally, determining a second target particle group to be optimized for the next iterative update based on the particle count includes: determining a first target particle group to be optimized whose particle count is less than a first preset threshold as a particle group to be deleted, deleting the particle group to be deleted, and updating the group count, wherein the first preset threshold is used to indicate the minimum number of particles to be optimized contained in each first target particle group to be optimized, and the group count is used to indicate the number of remaining first target particle groups to be optimized; when the group count is less than a second preset threshold and the number of times the second update method is executed is zero, determining the first target particle group to be optimized with the new particles added as the second target particle group to be optimized; when the group count is less than the second preset threshold and the number of times the second update method is executed is not zero, or when the group count is greater than or equal to the second preset threshold, determining multiple first target particle groups to be optimized after this iterative update as the second target particle group to be optimized.

[0014] According to another aspect of an embodiment of the present application, a device for determining an optimal parameter value is also provided, including: an acquisition module, used to acquire an equivalent circuit model of a battery to be tested, wherein the equivalent circuit model is used to indicate the change state of electrical parameters of the battery to be tested during the charging and discharging process, and during the charging and discharging process, the state of the battery to be tested includes: an electrochemical polarization state and a concentration difference polarization state, and the electrical parameters include: an inherent resistance of the battery to be tested, an open circuit voltage of the battery to be tested, a first resistance and a first capacitance of the battery to be tested in the electrochemical polarization state, and a second resistance and a second capacitance of the battery to be tested in the concentration difference polarization state; a determination module, used to obtain an equivalent circuit model of a battery to be tested, wherein the equivalent circuit model is used to indicate the change state of electrical parameters of the battery to be tested during the charging and discharging process, and the state of the battery to be tested includes: an electrochemical polarization state and a concentration difference polarization state, and the electrical parameters include: an inherent resistance of the battery to be tested, an open circuit voltage of the battery to be tested, a first resistance and a first capacitance of the battery to be tested in the electrochemical polarization state, and a second resistance and a second capacitance of the battery to be tested in the concentration difference polarization state; and a determination module, used to obtain an equivalent circuit model of a battery to be tested, wherein the equivalent circuit model is used to indicate the change state of electrical parameters of the battery to be tested during the charging and discharging process. A state space model of the battery to be tested is determined according to the electrical parameters, wherein the state space model is used to indicate the correlation between the input current of the battery to be tested, the output voltage of the battery to be tested and the electrical parameters; an analysis module is used to analyze the state space model using a particle swarm optimization algorithm to obtain an optimal parameter value, wherein the particle swarm optimization algorithm incorporates complex-order operators, the optimal parameter value is the value of the electrical parameter that minimizes the error value, the error value is the error value between the current measured voltage value and the current predicted voltage value of the battery to be tested at the current moment, and the current predicted voltage value is a voltage value determined according to the equivalent circuit model and the optimal parameter value.

[0015] According to another aspect of an embodiment of the present application, a non-volatile storage medium is provided, in which a computer program is stored, wherein the method for determining the optimal parameter value is executed by running the computer program in a device where the non-volatile storage medium is located.

[0016] According to another aspect of an embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the above-mentioned method for determining an optimal parameter value through the computer program.

[0017] According to another aspect of an embodiment of the present application, a computer program product is also provided, including computer instructions, which implement the steps of the above-mentioned method for determining optimal parameter values ​​when executed by a processor.

[0018] In an embodiment of the present application, an equivalent circuit model of a battery to be tested is obtained, wherein the equivalent circuit model is used to indicate the changing state of electrical parameters of the battery to be tested during the charging and discharging process. During the charging and discharging process, the state of the battery to be tested includes: electrochemical polarization state and concentration difference polarization state, and the electrical parameters include: inherent resistance of the battery to be tested, open circuit voltage of the battery to be tested, first resistance and first capacitance of the battery to be tested under electrochemical polarization state, second resistance and second capacitance of the battery to be tested under concentration difference polarization state; a state space model of the battery to be tested is determined according to the electrical parameters, wherein the state space model is used to indicate the correlation between the input current of the battery to be tested, the output voltage of the battery to be tested and the electrical parameters; a particle swarm optimization algorithm is used to analyze the state space model to obtain the optimal parameter value, wherein the ... Complex-order operators are integrated into the subgroup optimization algorithm. The optimal parameter value is the value of the electrical parameter that minimizes the error value. The error value is the error value between the current measured voltage value and the current predicted voltage value of the battery to be tested at the current moment. The current predicted voltage value is a voltage value determined according to the equivalent circuit model and the optimal parameter value. By integrating complex-order operators on the basis of the fractional-order method for identifying battery model parameters, the complex-order adaptation law is applied to the particle velocity adaptation mechanism of the algorithm, thereby achieving the purpose of fully utilizing the historical state information of the particle velocity, thereby achieving the technical effect of improving the algorithm's optimization ability and improving the accuracy of the identification results, and further solving the technical problem of the inability to accurately identify the parameters of the battery model caused by the low precision and poor robustness of the battery model parameter identification method in the related technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 is a hardware structure block diagram of a computer terminal for implementing a method for determining an optimal parameter value according to an embodiment of the present application;

[0021] Figure 2 is a flowchart of a method for determining an optimal parameter value according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of a fractional second-order resistance-capacitance (RC) effective circuit model according to an embodiment of the present application;

[0023] Figure 4is a pseudo code schematic diagram of iterative optimization using a complex-order Darwin particle swarm optimization algorithm according to an embodiment of the present application;

[0024] Figure 5 is a structural schematic diagram of a device for determining an optimal parameter value according to an embodiment of the present application;

[0025] Figure 6 is a parameter identification result record table according to an embodiment of the present application;

[0026] Figure 7 This is a schematic diagram of current-voltage changes of a lithium cobalt oxide manganese oxide battery under test conditions according to an embodiment of the present application;

[0027] Figure 8 It is a comparison diagram of the fitting value evolution curves of a complex-order Darwin particle swarm optimization algorithm (CoDPSO) and a particle swarm optimization algorithm (PSO) according to an embodiment of the present application;

[0028] Fig. 9 It is a comparison diagram of output voltage identification results of a complex-order Darwin particle swarm optimization algorithm (CoDPSO) and a particle swarm optimization algorithm (PSO) according to an embodiment of the present application;

[0029] Fig.10 is a comparison diagram of the absolute error of the identification result of a complex-order Darwin particle swarm optimization algorithm (CoDPSO) and the absolute error of the identification result of a particle swarm optimization algorithm (PSO) according to an embodiment of the present application;

[0030] Fig.11 It is a comparison diagram of the relative error of the identification result of a complex-order Darwin particle swarm optimization algorithm (CoDPSO) and the relative error of the identification result of a particle swarm optimization algorithm (PSO) according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] In order to better understand the embodiments of the present application, the technical terms involved in the embodiments of the present application are explained as follows:

[0034] Electrochemical polarization state: The polarization phenomenon that occurs inside a material under the action of an electric field. The electrochemical polarization state can be characterized by measuring the dielectric constant or capacitance of the material.

[0035] Concentration difference polarization state: refers to the polarization phenomenon caused by uneven electrolyte concentration in the battery. During the operation of the battery, due to the limited migration and diffusion speed of the electrolyte, it is easy to form a concentration gradient on the electrode surface, which is concentration difference polarization.

[0036] State of charge (SOC): Indicates the charging state of the battery, usually used to describe the ratio of the remaining battery power or the charged power to the rated capacity, that is, the degree of charge of the battery.

[0037] Open circuit voltage (OCV): The difference between the positive electrode potential and the negative electrode potential when the battery is open, which is approximately the electromotive force of the battery.

[0038] In the related art, the method for parameter identification of the battery model has the problems of low accuracy and poor robustness. The embodiments of the present application provide a related solution, which is described in detail below.

[0039] According to an embodiment of the present application, a method embodiment for determining optimal parameter values ​​is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0040] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1FIG. 1 shows a hardware structure block diagram of a computer terminal for implementing a method for determining an optimal parameter value. Figure 1 As shown, the computer terminal 10 may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0041] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10. As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0042] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage devices corresponding to the method for determining the optimal parameter value in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the above-mentioned method for determining the optimal parameter value is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0043] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0044] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 .

[0045] The embodiment of the present application provides a method for determining the optimal parameter value that can be run in the above-mentioned operating environment. Figure 2 is a flowchart of the steps of the method for determining the optimal parameter value provided in an embodiment of the present application, such as Figure 2 As shown, the method comprises the following steps:

[0046] Step S202, obtaining an equivalent circuit model of the battery to be tested, wherein the equivalent circuit model is used to indicate the change state of electrical parameters of the battery to be tested during the charging and discharging process. During the charging and discharging process, the state of the battery to be tested includes: electrochemical polarization state and concentration difference polarization state, and the electrical parameters include: inherent resistance of the battery to be tested, open circuit voltage of the battery to be tested, first resistance and first capacitance of the battery to be tested in the electrochemical polarization state, and second resistance and second capacitance of the battery to be tested in the concentration difference polarization state.

[0047] The battery equivalent circuit model parameter identification method based on the particle swarm optimization algorithm used in the related art has the problems of low accuracy and poor robustness. Therefore, the embodiment of the present application provides an improved model parameterization method based on the particle swarm optimization algorithm. In step S202, the equivalent circuit model of the battery to be parameter identified (i.e., the battery to be tested) is first obtained. By performing charge and discharge tests on the equivalent circuit model, the change state of the electrical parameters of the battery to be tested under the charge and discharge state can be simulated. In the embodiment of the present application, the research is mainly based on the following two states that exist in the battery under the charge and discharge state: electrochemical polarization state and concentration difference polarization state; wherein the electrical parameters observed during the research process are the parameters contained in the equivalent circuit model; Figure 3 is a schematic diagram of a fractional second-order resistor-capacitor (RC) equivalent circuit model. When the equivalent circuit model of the battery to be tested is as follows Figure 3When the second-order fractional second-order RC equivalent circuit model is shown, the electrical parameters are: electrochemical polarization resistance (i.e., the first resistance, R1) and electrochemical polarization capacitance (i.e., the first capacitance, C1), concentration difference polarization resistance (i.e., the second resistance R2), concentration difference polarization capacitance (C2), and the inherent ohmic resistance (R0) of the battery to be tested, such as Figure 3 As shown, the battery to be tested also includes the following electrical parameters: open circuit voltage OCV (the voltage value output when the battery to be tested is not connected to any load), current (I), terminal voltage V0 (voltage at the end of the circuit), and constant phase elements (CPE1, CPE2); in the second-order fractional second-order RC equivalent circuit model, the negative electrode of the terminal voltage V0 is used as the input end, and the input end is connected to the ohmic internal resistance R0 after passing through the open circuit voltage OCV, and the other end of the ohmic internal resistance R0 is connected in series with the loop formed by the electrochemical polarization resistor R1 and the capacitor C1, the output end of the loop of the resistor R1 and the capacitor C1 is connected to the input end of the loop formed by the concentration difference polarization resistor R2 and the capacitor C2, and the output end of the loop of the resistor R2 and the capacitor C2 is connected to the positive electrode of the terminal voltage V0.

[0048] Step S204, determining a state space model of the battery to be tested according to the electrical parameters, wherein the state space model is used to indicate a correlation between an input current of the battery to be tested, an output voltage of the battery to be tested, and the electrical parameters.

[0049] In step S204, a state space model of the battery to be tested is established based on the electrical parameters observed in step S202 that can reflect the changes in the electrochemical polarization state of the battery to be tested during the charging and discharging process and the electrical parameters that can reflect the changes in the concentration difference polarization state of the battery to be tested during the charging and discharging process, wherein the established state space model can reflect the mathematical (correlation) between the input current and input voltage of the battery to be tested and the above-mentioned electrical parameters observed in step S202.

[0050] Optionally, determining the state space model of the battery to be tested according to the electrical parameters includes: determining the current value of the battery to be tested at the current moment; determining the first voltage according to the first resistance, the first capacitance and the current value, and determining the second voltage according to the second resistance and the second capacitance, wherein the first voltage is used to indicate the electrochemical polarization effect of the battery to be tested during the charging and discharging process, and the second voltage is used to indicate the concentration range effect of the battery to be tested during the charging and discharging process; obtaining the state of charge SOC of the battery to be tested at the current moment, and determining the state variables of the state space model according to the SOC, the first voltage and the second voltage; determining an objective function for indicating the correlation between the SOC and the open circuit voltage of the battery to be tested, and determining the output variables of the state space model according to the objective function, the first voltage, the second voltage, the current value and the inherent resistance; determining the current value as the input variable of the state space model, and determining the state space model according to the state variables, the output variables and the input variables.

[0051] In this embodiment, the state space model can reflect the dynamic changes of the battery under test during the charging and discharging process, such as Figure 3 As shown, the fractional second-order RC equivalent circuit model of the battery to be tested includes two branches. During the charge and discharge process of the battery to be tested, current flows through both branches. In this embodiment, the state space model can reflect the internal dynamic changes of the battery to be tested during the charge and discharge process. When establishing the state space model, it is necessary to consider the current I(t) flowing through the two branches at the same time. According to Kirchhoff's law, it can be known that the parameters in the two branches in the equivalent circuit model have the following functional relationship Among them, V1 ′ represents the derivative of V1, V1 is Figure 3 The voltage of branch 1, V1, can reflect the change of electrochemical polarization state of the battery under test during the charge and discharge process; V2 ′ represents the derivative of V2, V2 is Figure 3 The change of the voltage V2 of the middle branch 2 can reflect the change of the concentration difference polarization state of the battery under test during the charge and discharge process; based on the above formula, the (first) voltage V1 of the battery under test in the electrochemical polarization state and the (second) voltage V2 of the battery under test in the concentration difference polarization state can be determined. Furthermore, the state of charge (SOC) of the battery under test will also change during the charge and discharge process. The change of the SOC of the battery under test over time is shown in the formula: Where η is the Coulomb efficiency of the battery to be tested, Q N is the nominal capacity of the battery to be tested, represents the derivative of SOC with respect to time; therefore, when establishing the state space model, the SOC of the battery to be tested needs to be taken into account at the same time. In this embodiment, the factor (i.e., the state variable x(t)) as the state variable in the state space model is determined according to multiple parameters (V1, V2, and SOC) that change during the charging and discharging process of the battery to be tested. x(t) ′ =Ax(t)+Bu(t), where u(t) represents the input variable of the state-space model (i.e., current I(t)), and A and B are the parameters of the battery to be tested (R1, C1, R2, C2, η, Q N ) is used to represent the state change of the battery under test during the charging and discharging process, for example, In addition, the voltage output by the battery under test during the charging and discharging process will also change. The output of the battery under test during the charging and discharging process is related to the initial open circuit voltage (OCV), state of charge (SOC) and other electrical parameters of the battery under test. In this embodiment, the output variable y(t) is determined based on the above parameters, where y(t)=Cx(t)+Du(t), C and D are variable factors determined according to the parameters of the battery under test (OCV, SOC, R0) and used to represent the output of the battery under test during the charging and discharging process. For example, D = -R0, where OCV(t) represents the open circuit voltage of the battery under test at different times t, and SOC(t) represents the state of charge of the battery under test at different times t. There is a strong nonlinear relationship between OCV and SOC, and the relationship between the two can be expressed as follows: OCV = a0 + a1SOC + a2SOC 2 +a3SOC 3 +a4SOC 4 , a0~a4 represent multiple coefficients. After the input variable I(t) (i.e. u(t)), state variable x(t) and output variable y(t) of the state space model are determined according to the above method, the state space model of the battery to be tested can be established, wherein the mathematical form of the state space model of the battery to be tested is as follows: Where η = [αβ1] T , is a transposed matrix, α and β are two fractional orders, α and β are determined according to the impedance change of the constant phase elements (CPE1 and CPE2) in the battery under test during the charge and discharge process of the battery under test. The change of the impedance of CPE1 during the charge and discharge process of the battery under test can be expressed by the following formula The change in the impedance of CPE2 during the charge and discharge process of the battery under test can be expressed by the following formula Where Z1 and Z2 represent complex impedances, s represents a complex variable, α and β are the fractional orders of the phase elements (CPE1 and CPE2); x(t) = [V1 V2 SOC] T is a transposed matrix, y(t) represents the output V0.

[0052] Step S206, using a particle swarm optimization algorithm to analyze the state space model to obtain an optimal parameter value, wherein the particle swarm optimization algorithm incorporates complex-order operators, and the optimal parameter value is the value of the electrical parameter that minimizes the error value, and the error value is the error value between the current measured voltage value of the battery to be tested at the current moment and the current predicted voltage value, and the current predicted voltage value is a voltage value determined based on the equivalent circuit model and the optimal parameter value.

[0053] In the embodiment of the present application, an improved Darwin particle swarm optimization algorithm (DPSO) is used to perform parameter identification on the state space model of the battery to be tested, and the optimal operating parameters (i.e., optimal parameter values) of the battery to be tested are determined. In the embodiment of the present application, the Darwin particle swarm optimization algorithm (DPSO) is improved by incorporating complex-order operators into the Darwin particle swarm optimization algorithm (DPSO). The improved Darwin particle swarm optimization algorithm (DPSO) is called a complex-order Darwin particle swarm optimization algorithm (CoDPSO). Due to the introduction of complex-order operators, the historical state of the battery to be tested can be fully considered when performing parameter identification (compared to the prior art, at least more consideration is given to the change in the frequency response characteristics of the capacitor during the power-on operation of the battery to be tested). In step S206, the particle swarm optimization algorithm (i.e. The CoDPSO (CoDPSO) analyzes the state space model of the battery to be tested established in step S204. During the analysis process, a set of electrical parameter values ​​that minimize the error between the current measured voltage value and the predicted voltage value of the battery to be tested at the current moment is determined as the optimal operating parameter (i.e., the optimal parameter value) of the battery to be tested; the predicted voltage value in the above scheme is a voltage value jointly determined by the optimal parameter value determined by the complex-order Darwin particle swarm optimization algorithm (CoDPSO) and the equivalent circuit model of the battery to be tested; the method for determining the predicted voltage value is as follows: the optimal parameter value is substituted into the equivalent circuit model. At this time, each electrical parameter in the equivalent circuit model can find a corresponding value in the optimal parameter value. By substituting it into the equivalent circuit model, the voltage (i.e., the predicted voltage) when the value of the electrical parameter is the optimal parameter value can be determined according to the equivalent circuit model. When the battery to be tested operates at the optimal parameter value, it can effectively improve the charging and discharging efficiency of the battery, so that the battery can convert energy more efficiently during the charging and discharging process and reduce energy loss; it can effectively extend the life of the battery: by optimizing the parameter value, the internal resistance of the battery can be reduced, the loss of the battery can be reduced, and the service life of the battery can be extended; it can also improve the stability of the battery: the optimal parameter value can enable the battery to maintain stable performance under various working environments, and improve the reliability and safety of the battery. Therefore, after the optimal parameter value is determined using the method provided in this embodiment, the adjustable parameters in the battery can be adjusted to the optimal parameter value to make it operate more efficiently and stably.

[0054] In step S206, the neural network executing the complex-order Darwin particle swarm optimization algorithm (CoDPSO) may be loaded into the memory, for example, the raw data of the neural network may be loaded from the non-volatile memory into the volatile memory so that the processor runs the neural network. The raw data of the neural network refers to unprocessed data, which generally includes parameters and structural data of the neural network. The structural data may be a calculation relationship based on the parameters, such as a forward propagation calculation relationship between intermediate layers and neurons. Specifically, the structural data may include codes related to the structure of the neural network, such as codes for performing related calculations between intermediate layers and neurons.

[0055] In one embodiment, an area for loading a neural network for executing a complex-order Darwin particle swarm optimization algorithm (CoDPSO) can be divided in the memory, which may include a structure data storage area and a parameter storage area. The structure data storage area is used to store structure-related codes, and the parameters referenced by it can point to the address of a specific parameter in the parameter storage area through a pointer. During the training process of the neural network, it may be necessary to frequently update the parameters, and the parameter values ​​in the parameter storage area can be updated.

[0056] According to an optional embodiment of the present application, a particle swarm optimization algorithm is used to analyze the state space model to obtain the optimal parameter value, including: obtaining multiple groups of historical state information of the state space model at multiple historical moments before the current moment, and generating multiple first target particle swarms to be optimized based on the multiple groups of historical state information, wherein each first target particle swarm to be optimized contains multiple particles to be optimized, each group of historical state information is a particle to be optimized, each historical moment corresponds to a group of historical state information, each group of historical state information at least includes: the historical position of the particle to be optimized at multiple historical moments before the current moment, the historical position is used to indicate the value of the particle to be optimized at the historical moment, each particle to be optimized includes: the value of the first type of parameters contained in the state variables in the state space model and the value of the second type of parameters contained in the output variables in the state space model, wherein the first type of parameters includes: a first capacitor, a first resistor, a second capacitor and a second resistor, the values ​​of the first type of parameters include: a capacitance value of the first capacitor, a first fractional order of the first capacitor, a resistance value of the first resistor, a capacitance value of the second capacitor, a second fractional order of the second capacitor and a resistance value of the second resistor, the first fractional order is used to indicate a frequency response characteristic of the first capacitor, and the second fractional order is used to indicate a frequency response characteristic of the second capacitor, the second type of parameters include: a plurality of coefficients in an objective function for generating an output variable, and the values ​​of the second type of parameters include: a plurality of values ​​corresponding to the plurality of coefficients; a particle swarm optimization algorithm is used to iteratively update a plurality of first target particle swarms to be optimized until the number of iterations reaches a preset number of iterations, and a target particle swarm is output, the target particle swarm being the result of the last iterative update; an error value corresponding to each particle in the target particle swarm is determined, a particle corresponding to a minimum error value is determined as a target particle, and a plurality of values ​​contained in the target particle are determined as optimal parameter values.

[0057] In this embodiment, a complex-order Darwin particle swarm optimization algorithm (CoDPSO) is used to analyze the variables (state variable x(t), output variable y(t)) in the state space model of the battery to be tested to determine the optimal operating parameters (i.e., optimal parameter values) of the battery to be tested, wherein, during the analysis process, the complex-order Darwin particle swarm optimization algorithm (CoDPSO) performs iterative optimization based on the historical state information of the state space model to obtain the optimal parameter value; before the iterative optimization, firstly obtain multiple groups of historical state information of the state space, wherein the historical state information is the state of the state space model at multiple historical moments before the current moment; in this embodiment, the state information of the state space model at each historical moment (i.e., historical state information) is taken as a particle to be optimized, and the set composed of multiple groups of historical state information is regarded as a cluster to be iteratively optimized (i.e., the first target particle group to be optimized); in this embodiment, multiple particles to be optimized are divided into multiple clusters, that is, there are multiple first target particle groups to be optimized; before starting the iterative optimization, a first target particle group to be optimized may contain one particle to be optimized or multiple particles to be optimized. It can be seen from the above embodiments that the state space model includes multiple variables (such as state variable x(t), output variable y(t), etc.), and the multiple variables included in the state space model are generated according to the basic electrical parameters of the battery to be tested (such as R0, R1, R2, C1, C2), the complex order operator (α and β) that affects the impedance of the battery to be tested during the charging and discharging process, and multiple (second type) parameters (a0, a1, a2, a3, a4) that affect the output of the battery to be tested. Therefore, the state of the state space model at each moment is actually is the state of the above multiple parameters (R0, R1, R2, C1, C2, α, β and a0, a1, a2, a3, a4) at this moment. Therefore, each set of historical state information of the state space model at each historical moment can be expressed as a set of parameter values ​​(R0, R1, R2, C1, C2, α, β and a0, a1, a2, a3, a4) at this historical moment. This set of parameters (R0, R1, R2, C1, C2, α, β and a0, a1, a2, a3, a4) is a particle to be optimized. In the iterative optimization process, the parameter values ​​of the above multiple parameters at each historical moment are regarded as the positions of the above multiple parameters at each historical moment (i.e., historical positions).According to the above, the process of using the complex-order Darwin particle swarm optimization algorithm (CoDPSO) to analyze the state space model of the battery to be tested is the process of using the complex-order Darwin particle swarm optimization algorithm (CoDPSO) to determine the numerical values ​​of the (first type) parameters (R0, R1, R2, C1, C2, α, β) of the battery to be tested and the numerical values ​​of the (second type) parameters (a0, a1, a2, a3, a4) related to the output of the battery to be tested, α is a complex-order operator, and is also a physical quantity used to indicate the frequency response characteristics of the capacitor C1, β is a complex-order operator, and is also a physical quantity used to indicate the frequency response characteristics of the capacitor C2. Through the above method, the problem of analyzing the operating state of the battery to be tested is converted into the problem of determining the numerical values ​​of the parameters to be identified (R0, R1, R2, C1, C2, α, β and a0, a1, a2, a3, a4). When the number of iterative optimizations for the particle cluster (i.e., the first target particle cluster to be optimized) reaches the preset number of iterative optimizations, the above parameters (R1, R2, C1, C2, α, β and a0, a1, a2, a3, a4) are output with the numerical values ​​after the last iterative optimization is completed. The output may be a set of parameter values ​​of the above parameters, or may be multiple sets of parameter values ​​of the above parameters, that is, the target particle cluster may contain a set of parameter values ​​of the above parameters or may contain multiple sets of parameter values ​​of the above parameters; next, a set of parameter values ​​(i.e., one particle) with the smallest error value is determined in the target particle cluster as the target particle, and it is output as the optimal parameter value.

[0058] Optionally, a particle swarm optimization algorithm is used to iteratively update multiple first target particle swarms to be optimized, including: for each first target particle swarm to be optimized, determining an individual optimal position and a group optimal position according to historical state information, the individual optimal position is the historical position of each particle to be optimized when the historical error value is the smallest, the historical error value is the root mean square difference between multiple predicted voltage values ​​and multiple measured voltage values ​​corresponding to multiple historical moments, and the group optimal position is determined from multiple individual optimal positions; updating the speed and position of each particle to be optimized according to the individual optimal position and the group optimal position to obtain a speed update result and a position update result, wherein the speed update result is an adjustment direction and an adjustment value of the particle to be optimized when performing iterative update, and the adjustment direction includes: increase and decrease, position The update result is set to be the value of the particles to be optimized after iterative update; the update method of the first target particle group to be optimized is determined according to the speed update result and the position update result, and the first target particle group to be optimized is updated according to the update method, and the update method includes: a first update method and a second update method, the first update method is used to indicate to increase the number of particles to be optimized contained in the first target particle group to be optimized, and the second update method is used to indicate to reduce the number of particles to be optimized contained in the first target particle group to be optimized; determine the particle count of each first target particle group to be optimized, and determine the second target particle group to be optimized for the next iterative update according to the particle count, wherein the particle count is the number of particles to be optimized contained in the first target particle group to be optimized after being updated according to the update method.

[0059] The main difference between CoDPSO and the classic particle swarm optimization algorithm (PSO) is that the population of CoDPSO is composed of multiple parallel particle swarms. In CoDPSO, each particle swarm runs the optimization process of PSO. The collection of particle swarms is subject to the natural selection mechanism (that is, Darwin's principle of survival of the fittest), which enhances the ability of PSO to get rid of local optimality. The principle is to run multiple PSO algorithms on the same search problem at the same time and apply the natural selection mechanism to improve the search ability of the algorithm. When the search tends to the local optimum, the search in this area is abandoned and a new search is started in another area. In this method, each step rewards the group with better results by extending the life of particles or generating new offspring, while the group with poor results is punished by reducing the life of the group or deleting particles. In order to analyze the overall state of each group, the solutions of all particles are evaluated, and the global optimal solution of the given group and the individual local optimal position of each particle in the group are updated. If a new global solution is found, a new particle is generated. If a particle group fails to find a better solution within the specified number of steps, it will be removed. In this embodiment, the iteration process is performed simultaneously on multiple clusters to be optimized (ie, the first target particle cluster to be optimized) and new particle clusters that may be generated during the iteration process. Figure 4 This is a pseudo code diagram of iterative optimization using the complex-order Darwin particle swarm optimization algorithm, such as Figure 4 As shown, each time the iterative optimization is performed, the individual best position is determined and the optimal position of the group Among them, k represents the number of clusters to be optimized (i.e., the first target particle group to be optimized), i represents the i-th particle; the individual best position There are multiple, one is determined in each cluster to be optimized; for each cluster to be optimized, it contains multiple particles (i.e. multiple groups of historical states, multiple groups of parameter values), and the fitness values ​​(i.e., fit values) are calculated based on these multiple groups of parameter values, and the group of parameter values ​​corresponding to the fitness value with the smallest value is determined as the individual best position of the cluster to be optimized; the best position Only one is determined among all particles (i.e., all historical states, all parameter values) contained in all clusters to be optimized, and it is the particle that can minimize the fitness value. In this embodiment, since the individual best positions in each cluster have been determined, the fitness value can be calculated again based on these individual best positions, and the individual best position corresponding to the fitness value with the smallest value is determined as the group best position. The above fitness value is calculated according to the following formula: Wherein, E represents the fitness value (i.e., historical error value, fitting value), k represents the kth iteration, M represents the total number of iterations, V0(k) is the actual voltage value (i.e., measured voltage value) measured for the battery under test at the measurement time corresponding to the kth iteration, is the estimated voltage, which is the predicted voltage calculated using the result of the k-th iteration of the particle i to be optimized (i.e., a set of parameter values ​​after completing the k-th iteration). Figure 4 As shown, after the individual optimal position and the group optimal position are determined in the cluster to be optimized by the above method, the speed and position (i.e., value) of the particle are updated based on the individual optimal position and the group optimal position to obtain an update result, wherein the update result includes two results: a speed update result and a position update result. In this embodiment, a particle is a set of parameters to be identified (R0, R1, R2, C1, C2, α, β and a0, a1, a2, a3, a4), and the particle speed update result represents the adjustment direction (increase or decrease) and specific adjustment value of the value of this set of parameters in the next iteration. In the scenario described in this embodiment where each particle to be optimized contains multiple parameters, the speed update result is a multidimensional vector, each component represents the adjustment direction and adjustment value of one of the parameters, and the position update result is the value of each parameter after the speed update result is updated. Further, after the speed and position of the particle are updated, the cluster to be optimized is optimized, such as Figure 4 As shown, there are two types of update methods, one is to update the cluster to be optimized by generating new particles (i.e., the first update method), and the other is to update the cluster to be optimized by reducing the number of particles to be optimized contained in the cluster to be optimized (i.e., the second update method). In this embodiment, the second update method is implemented by deleting the worst particle, wherein the worst particle is a particle to be optimized (i.e., a set of parameter values) corresponding to the fitness value with the largest value when calculating the fitness value. After each cluster to be optimized is updated by one of the methods, the number of particles to be optimized (i.e., particle count N) contained in each cluster (including the cluster to be optimized initially used for iteration and the new cluster that may be generated during the update process) after the update process is completed is determined, and the particle cluster to participate in the iteration next time (i.e., the second target particle cluster to be optimized) is determined according to the particle count.

[0060] According to an optional embodiment of the present application, the speed and position of each particle to be optimized are updated according to the individual optimal position and the group optimal position, including: updating the speed of each particle to be optimized according to the following formula: in, represents the complex order operator, α1 represents the first fractional order, β1 represents the second fractional order, represents the speed update result of particle i to be optimized after the kth iteration update, k represents the number of iterations, c1 and c2 are different preset values, r1 and r2 are random numbers within the preset value range, represents the individual best position of particle i to be optimized during the k-th iteration update, represents the position of particle i to be optimized before the kth iteration update, Represents the optimal position of the group during the k-th iteration update.

[0061] As mentioned in the above embodiment, in each iteration, after determining the individual best position and the group best position in the cluster to be optimized, the speed and position of the particle to be optimized are updated according to the individual best position and the group best position, wherein the speed is used to indicate the change (increase / decrease, and the specific value of increase / decrease) of the particle in this iteration during this iteration, which is a variable; the position is used to indicate the value of the particle to be optimized after this iteration, which is affected by the speed. Specifically, in this embodiment, the speed of the particle to be optimized is updated according to the following formula: in, Update results for speed, It is a complex-order operator determined according to the (first) fractional order α1 of the capacitor C1 and the (second) fractional order β1 of the capacitor C2 in the equivalent circuit model; α1 and α in the above embodiment are both the (first) fractional order of the capacitor C1, and β1 and β in the above embodiment are both the (second) fractional order of the capacitor C2, but because the frequency response characteristics of the capacitor will change during the operation of the battery to be tested, α1 and α, β1 and β are slightly different. In the above speed update formula, k represents the number of iterations, c1 and c2 are different preset values, which are common operators in the Darwin particle swarm optimization algorithm, representing the weights of cognitive and social components, and reflecting the memory of the optimized particles of the entire group's historical best position; r1 and r2 are positive random numbers between the interval [0,1] (i.e., the preset range); represents the position of particle i to be optimized at the beginning of the kth iteration (i.e., the parameter value at the beginning of the kth iteration), represents the individual best position of the particle i to be optimized after the k-th iteration update (it is the individual best position of the particle i to be optimized so far before the start of the k-th update), represents the optimal position of the group when performing the kth iteration update (it is the optimal position of the group to be optimized participating in this iteration before starting the kth update). According to the above method, we can know that the complex-order operator Affects the velocity update of the particle to be optimized. If the complex-order operator Performing Taylor expansion, the velocity update formula can also be expressed as follows: in, Represents the velocity of particle i to be optimized at the (k-2)th iteration.

[0062] According to another optional embodiment of the present application, updating the speed and position of each particle to be optimized according to the individual optimal position and the group optimal position also includes: updating the position of each particle to be optimized according to the following formula: Where k represents the number of iterations, represents the position update result of particle i to be optimized after the kth iteration update, represents the position of particle i to be optimized during the k-th iteration update, Represents the current velocity of particle i to be optimized during the k-th iterative update.

[0063] As mentioned in the above embodiment, the position update result is affected by the speed update result. In this embodiment, using Represents the process of updating the position of the particle to be optimized each time iterates; Represents the position of particle i to be optimized after the k-th iteration update (i.e., the position update result), and is also the initial position of the particle to be optimized at the beginning of the (K+1)-th iteration; represents the position of particle i to be optimized before the kth iteration update, Represents the current speed of the particle i to be optimized when the kth iterative update is performed. According to the speed update formula and the position update formula in the above embodiment, it can be seen that the speed update of the particle (to be optimized) depends not only on the previous speed value of the particle, but also on the historical information of the speed value. The speed update of the particle depends on short-term memory. To a certain extent, the diversity and distribution of the population can be protected; so that multiple clusters to be optimized can be taken into account at the same time during the update process; in addition, according to the speed update formula and the position update formula in the above embodiment, it can be seen that the speed update result at each moment depends on the speed value of the three moments before this moment; therefore, the method (CoDPSO) adopted in the embodiment of the present application can adjust the optimization direction according to the historical trajectory of the particle to be optimized; it helps to improve the local search accuracy of the algorithm, so that the particle to be optimized converges to the optimal solution faster, and improves the reliability of the iterative result and the stability of the iterative process.

[0064] According to some optional embodiments of the present application, a method for updating the first target particle swarm to be optimized is determined according to the speed update result and the position update result, including: determining the error value corresponding to each particle to be optimized after this iterative update according to the following formula: Wherein, E represents the error value, k represents the kth iteration update, M represents the total number of iteration updates, V0(k) is the measured voltage value of the battery under test when executing the kth iteration update, is a voltage value determined according to the equivalent circuit model and the speed update result and the position update result obtained by the k-th iterative update of the particle i to be optimized; the error value is compared with the previous error value to obtain a comparison result, and an update method is determined according to the comparison result, wherein the previous error value is an error value determined according to the previous iterative update result of the particle to be optimized; when the comparison result indicates that the error value is less than the previous error value, the update method is determined to be the first update method; when the comparison result indicates that the error value is greater than or equal to the previous error value, the second update method is determined.

[0065] As mentioned in the above embodiment, the iterative update process includes two update methods: "updating the cluster to be optimized by generating new particles" (i.e., the first update method) and "updating the cluster to be optimized by reducing the number of particles to be optimized contained in the cluster to be optimized" (i.e., the second update method). Figure 4 As shown, whether the first updating method or the second updating method is specifically adopted is determined by the fitness value (i.e., error value) of the particle to be optimized. In this embodiment, the selection of which updating method is determined by comparing the new fitness value (the error value calculated at the beginning of this iteration) with the old fitness value (i.e., the previous error value). When the new fitness value is less than the old fitness value, the above-mentioned first updating method is adopted: generating new particles; when the new fitness value is greater than or equal to the old fitness value, the above-mentioned second updating method is adopted (deleting particles); in this embodiment, according to the formula: To calculate the fitness value (i.e., error value), where E represents the fitness value (i.e., fitness value, error value), k represents the kth iteration, M represents the total number of iterations, and V0(k) is the actual voltage value measured for the battery under test at the measurement time corresponding to the kth iteration. is the estimated voltage, which is the voltage value calculated using the result of the k-th iteration of the particle i to be optimized (i.e., a set of parameter values ​​after completing the k-th iteration); for example, if this update is the k-th update, then is the new error value mentioned above, is the old fit value (i.e., the previous error value), where V0(k-1) is the actual voltage value measured for the battery under test at the measurement time corresponding to the (k-1)th iteration update, It is the predicted voltage value calculated using the result of the (k-1)th iteration of particle i to be optimized (i.e., a set of parameter values ​​after completing the (k-1)th iteration).

[0066] According to an optional embodiment of the present application, updating the first target particle group to be optimized according to the updating method includes: updating the first target particle group to be optimized according to the first updating method, and updating the first target particle group to be optimized according to the second updating method, wherein updating the first target particle group to be optimized according to the first updating method includes: determining a data set consisting of a velocity update result and a position update result as a new particle; adding the new particle to the particle group to be optimized, and updating the particle count; updating the first target particle group to be optimized according to the second updating method includes: determining an error value corresponding to each particle to be optimized among a plurality of particles to be optimized currently contained in the first target particle group to be optimized; determining the particle to be optimized corresponding to the error value with the largest numerical value as the worst particle; deleting the worst particle, and updating the particle count.

[0067] When the first update method is used to update the cluster to be optimized, new particles need to be generated. In this embodiment, the particles adjusted according to the speed update results and the position update results (a set of new parameter values ​​of R0, R1, R2, C1, C2, α, β, a0, a1, a2, a3, and a4) are used as new particles; the new particles are added to the cluster to be optimized (i.e., the first target particle group to be optimized) that is being iterated, and the particle count N of the cluster to be optimized is updated, where N is determined according to the number of particles to be optimized contained in the cluster to be optimized. When the second update method is used to update the cluster to be optimized, the particles in the cluster to be optimized need to be deleted, and the same method is used to update the particle count N of the cluster to be optimized. Figure 4 As shown in FIG. 1 , in this embodiment, the particle deleted when the second updating method is adopted is the worst particle among the multiple particles to be optimized contained in the cluster to be optimized. The worst particle is the particle to be optimized corresponding to the largest error value after calculating the error value (i.e., the fitting value) for each particle to be optimized. Figure 4 As shown, after the cluster to be optimized is updated by the second updating method, the particle count N of the cluster to be optimized also needs to be updated. At the same time, the number of deleted particles to be optimized N also needs to be recorded. kill In some scenarios, you can also set execution conditions for the second update method to accelerate the convergence speed of the particles to be optimized; of course, if you do not set execution conditions for the second update method, it will not affect the iteration of the particles to be optimized. Figure 4 As shown in FIG. 1 , the number of clusters to be optimized (i.e., the number of times particles are deleted SC) can be compared with the preset maximum number of deletions SC. max "Equal" is set as the execution condition of the second update method. max Execute the second update method when the second update method is executed; if the execution condition is set for the second update method, then after the cluster to be optimized is updated using the second update method, it is necessary not only to update the particle count N of the cluster to be optimized, but also to record the number N of the particles to be optimized that have been deleted. killSC also needs to be reinitialized, where SC can be initialized to 0 or according to the formula: Initialize SC to

[0068] According to another optional embodiment of the present application, a second target particle group to be optimized for the next iterative update is determined based on the particle count, including: determining the first target particle group to be optimized whose particle count is less than the first preset threshold as the particle group to be deleted, deleting the particle group to be deleted, and updating the group count, wherein the first preset threshold is used to indicate the minimum number of particles to be optimized contained in each first target particle group to be optimized, and the group count is used to indicate the number of remaining first target particle groups to be optimized; when the group count is less than the second preset threshold and the number of times the second update method is executed is zero, the first target particle group to be optimized with the new particles added is determined as the second target particle group to be optimized; when the group count is less than the second preset threshold and the number of times the second update method is executed is not zero, or when the group count is greater than or equal to the second preset threshold, the multiple first target particle groups to be optimized after this iterative update are determined as the second target particle group to be optimized.

[0069] Each time the cluster to be optimized (i.e., the first target particle group to be optimized) is iteratively updated to obtain a new cluster to be optimized (i.e., the second target particle group to be optimized). In this embodiment, the generation of the new cluster to be optimized (i.e., the second target particle group to be optimized) may be by deleting some clusters to be optimized from the multiple clusters to be optimized (i.e., the first target particle group to be optimized), or by generating a new particle group (a collection of multiple new particles generated during the update process); Figure 4 As shown, according to the particle count N of the cluster to be optimized, whether to delete the particles to be optimized during the iteration process (it can be determined by whether the second update method has been executed during the update process), the total number of clusters to be optimized N S These three factors together determine whether the method of generating a new cluster to be optimized is 1) generating a new cluster to be optimized or 2) deleting one or more clusters to be optimized among multiple clusters to be optimized. Figure 4 As shown, if after the iteration is completed, the number of particles contained in the cluster to be optimized (i.e., the particle count N) is less than the preset particle count minimum value N min (i.e., the first preset threshold), a new cluster to be optimized (i.e., the second target particle group to be optimized) is generated by deleting the particle group; wherein, the particle group to be deleted (i.e., the particle group to be deleted) is the particle group containing the number of particles to be optimized less than N min(i.e., the first preset threshold) particle group, this particle group to be deleted may be a newly generated particle group during the iteration process, or it may be the result of the original cluster to be optimized (i.e., the first target particle group to be optimized) after iterative update. min (i.e., the first preset threshold) of the particle group (i.e., the particle group to be deleted), the total number of clusters to be optimized (i.e., the group count N S ), and according to N S Further determine whether to "take the set consisting of the new particle group generated in the iterative update process and the remaining clusters to be optimized (after deleting the particle group to be deleted) as the new cluster to be optimized (i.e., the second target particle group to be optimized)" or "take the remaining clusters to be optimized (after deleting the particle group to be deleted) as the new cluster to be optimized (i.e., the second target particle group to be optimized)". Specifically, Figure 4 As shown in the figure, when ① after deleting the particle group to be deleted, the number of remaining clusters to be optimized (i.e., the population count N) is satisfied. S ) is less than the preset group count maximum value N s max (i.e., the second preset threshold), and ② no particles to be optimized have been deleted during the update of the optimized cluster (i.e., N kill =0), the set of new particles generated in the iterative update process and the remaining clusters to be optimized (after deleting the particle group to be deleted) is used as the new cluster to be optimized (i.e., the second target particle group to be optimized); and the group count N is updated. S Otherwise, if the above conditions ① and ② cannot be met at the same time, the remaining cluster to be optimized (after deleting the particle group to be deleted) will be used as the new cluster to be optimized (i.e., the second target particle group to be optimized). Figure 4 As shown, after the cluster to be optimized (ie, the first target particle group to be optimized) is updated and a new cluster to be optimized (ie, the second target particle group to be optimized) is determined, a complete iteration is completed, the iteration count is updated, and an iteration ends.

[0070] Through the above steps, the complex-order adaptation law can be applied to the particle speed adaptation mechanism of the Darwin particle swarm optimization algorithm, thereby improving the algorithm's optimization ability; when identifying the parameters of the battery to be tested by the particle swarm optimization algorithm that incorporates the complex-order adaptation law, the historical state information of the battery to be tested is fully considered, thereby improving the accuracy of the parameter identification result and the robustness of the identification process; and accelerating the speed of finding the optimal parameter value in the parameter identification process.

[0071] Figure 5 is a schematic diagram of the structure of a device for determining the optimal parameter value, such as Figure 5As shown, the device for determining the optimal parameter value includes: an acquisition module 50, which is used to obtain an equivalent circuit model of the battery to be tested, wherein the equivalent circuit model is used to indicate the change state of the electrical parameters of the battery to be tested during the charging and discharging process. During the charging and discharging process, the state of the battery to be tested includes: an electrochemical polarization state and a concentration difference polarization state, and the electrical parameters include: an inherent resistance of the battery to be tested, an open circuit voltage of the battery to be tested, a first resistance and a first capacitance of the battery to be tested in the electrochemical polarization state, and a second resistance and a second capacitance of the battery to be tested in the concentration difference polarization state; a determination module 52, which is used to determine the battery to be tested according to the electrical parameters. A state space model of the battery to be tested, wherein the state space model is used to indicate the correlation between the input current of the battery to be tested, the output voltage of the battery to be tested and the electrical parameters; an analysis module 54, used to analyze the state space model using a particle swarm optimization algorithm to obtain an optimal parameter value, wherein the particle swarm optimization algorithm incorporates complex-order operators, the optimal parameter value is the value of the electrical parameter that minimizes the error value, the error value is the error value between the current measured voltage value and the current predicted voltage value of the battery to be tested at the current moment, and the current predicted voltage value is a voltage value determined based on the equivalent circuit model and the optimal parameter value.

[0072] It should be noted that Figure 5 The preferred implementation of the illustrated embodiment can be found in Figure 2 The relevant description of the illustrated embodiment will not be repeated here.

[0073] In order to verify the complex-order Darwin particle swarm optimization algorithm (CoDPSO) in the embodiment of the present application, the following comparative experiment was carried out. During the experiment, the test condition was: urban dynamic stress test (Dynamic Stress Test, DST) condition, and the cobalt oxide lithium manganese oxide battery (LNMC) was used as the battery to be tested. Under the above test conditions, CoDPSO and particle swarm optimization algorithm (Particle swarm optimization, PSO) were used to identify the parameters of the cobalt oxide lithium manganese oxide battery (LNMC). It can be seen from the above embodiments that the parameters to be identified include: the objective function established according to the open circuit voltage (OCV) and the state of charge (SOC) of the battery to be tested (LNMC): OCV = a0 + a1SOC + a2SOC 2 +a3SOC 3 +a4SOC 4The multiple coefficients (a0, a1, a2, a3, a4) in the LNMC, the multiple electrical parameters of the battery to be tested (the inherent resistance R0, electrochemical polarization resistance R1, electrochemical polarization capacitance C1, concentration difference polarization capacitance C2, concentration difference polarization resistance R2 of the battery to be tested), and other parameters related to the components in the battery to be tested: the fractional order α used to indicate the frequency response characteristics of the electrochemical polarization capacitance C1, the fractional order β used to indicate the frequency response characteristics of the concentration difference polarization capacitance C2, Figure 6 is a record table of parameter identification results. The CoDPSO algorithm is used to identify the parameters of the lithium cobalt oxide manganese oxide battery under the urban dynamic pressure test condition. The following results are obtained: Figure 6 The identification results are shown. Figure 7 This is a schematic diagram of the current-voltage change of the cobalt-cobalt lithium manganese acid battery under the test conditions. Under the entire test conditions, the current-voltage of the battery to be tested (cobalt-cobalt lithium manganese acid battery) is as follows Figure 7 The change curve shown; Figure 8 to Figure 11 Used to compare and verify the CoDPSO algorithm and PSO algorithm; Figure 8 This is a comparison of the fitting value evolution curves of the complex-order Darwin particle swarm optimization algorithm (CoDPSO) and the particle swarm optimization algorithm (PSO). Figure 8 The dashed line represents the PSO algorithm, and the solid line represents the CoDPSO algorithm; Figure 8 As shown in the figure, when the CoDPSO algorithm is used for parameter identification, the convergence speed is faster, that is, the optimal parameter value is obtained faster. Fig. 9 This is a comparison chart of the output voltage identification results of the complex-order Darwin particle swarm optimization algorithm (CoDPSO) and the particle swarm optimization algorithm (PSO). Fig. 9 In the figure, the dash-dot line and dotted line represent CoDPSO and PSO respectively; the black line represents the output voltage measurement value of the lithium cobalt oxide manganese oxide battery. Fig.10 This is a comparison chart of the absolute error of the identification results of the complex-order Darwin particle swarm optimization algorithm (CoDPSO) and the absolute error of the identification results of the particle swarm optimization algorithm (PSO). Fig.10 In the figure, the solid line represents the absolute error of the PSO algorithm, which is the difference between the voltage predicted by the PSO algorithm and the actual measured value of the cobalt-acid lithium manganese-acid battery at different times; the dotted line represents the absolute error of the CoDPSO algorithm, which is the difference between the voltage predicted by the CoDPSO algorithm and the actual measured value of the cobalt-acid lithium manganese-acid battery at different times; Fig.10 It can be seen that the maximum absolute error of the PSO algorithm is 0.0419V (volt), and the maximum absolute error of the CoDPSO algorithm is 0.0366V, that is, the identification result obtained by using the CoDPSO algorithm for parameter identification is more accurate, and the identification accuracy of the CoDPSO algorithm is higher. Fig.11This is a comparison chart of the relative error of the identification results of the complex-order Darwin particle swarm optimization algorithm (CoDPSO) and the relative error of the identification results of the particle swarm optimization algorithm (PSO). Fig.11 In the figure, the solid line represents the relative error of the PSO algorithm, which is the difference between the absolute error value of the PSO algorithm and the actual measured value of the cobalt-acid lithium manganese acid battery at different times; the dotted line represents the relative error of the CoDPSO algorithm, which is the difference between the absolute error value of the CoDPSO algorithm and the actual measured value of the cobalt-acid lithium manganese acid battery at different times; according to Fig.11 It can be seen that the maximum relative error of the PSO algorithm is 1.15%, and the maximum relative error of the CoDPSO algorithm is 0.96%. That is, the identification result obtained by using the CoDPSO algorithm for parameter identification is more accurate, the identification accuracy of the CoDPSO algorithm is higher, and the particle convergence speed in the identification process is faster. Furthermore, the accuracy of the two algorithms is evaluated according to the root mean square difference (RMSE) and the mean absolute error (MAE), wherein the smaller the values ​​of RMSE and MAE, the more accurate the prediction results and the higher the algorithm accuracy; according to the identification results of the CoDPSO algorithm in the experimental data and the actual measured voltage values, it is calculated that when the CoDPSO algorithm is used, the root mean square difference (RMSE) of the measurement result and the actual voltage value is 11.4, and the mean absolute error (MAE) is 9.2; according to the identification results of the PSO algorithm in the experimental data and the actual measured voltage values, when the PSO algorithm is used, the root mean square difference (RMSE) of the measurement result and the actual voltage value is 12.6, and the mean absolute error (MAE) is 10.1; it can be known that the CoDPSO algorithm provided in the embodiment of the present application has the advantages of high accuracy and fast convergence speed, and can achieve the technical effect of improving the accuracy of the identification results and reducing the time-consuming identification parameters.

[0074] An embodiment of the present application further provides a non-volatile storage medium, in which a computer program is stored, wherein the above method for determining the optimal parameter value is executed by running the computer program on a device where the non-volatile storage medium is located.

[0075] The non-volatile storage medium is used to store a program for performing the following functions: obtaining an equivalent circuit model of a battery to be tested, wherein the equivalent circuit model is used to indicate the changing state of electrical parameters of the battery to be tested during the charging and discharging process. During the charging and discharging process, the state of the battery to be tested includes: an electrochemical polarization state and a concentration difference polarization state, and the electrical parameters include: an inherent resistance of the battery to be tested, an open circuit voltage of the battery to be tested, a first resistance and a first capacitance of the battery to be tested in the electrochemical polarization state, and a second resistance and a second capacitance of the battery to be tested in the concentration difference polarization state; determining a state space model of the battery to be tested according to the electrical parameters, wherein the state space model is used to indicate the correlation between the input current of the battery to be tested, the output voltage of the battery to be tested and the electrical parameters; analyzing the state space model using a particle swarm optimization algorithm to obtain an optimal parameter value, wherein a complex-order operator is integrated into the particle swarm optimization algorithm, and the optimal parameter value is the value of the electrical parameter that minimizes the error value, and the error value is the error value between the current measured voltage value and the current predicted voltage value of the battery to be tested at the current moment, and the current predicted voltage value is a voltage value determined according to the equivalent circuit model and the optimal parameter value.

[0076] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the above method for determining an optimal parameter value through the computer program.

[0077] The processor in the above electronic device is used to run a program that performs the following functions: obtaining an equivalent circuit model of the battery to be tested, wherein the equivalent circuit model is used to indicate the change state of the electrical parameters of the battery to be tested during the charging and discharging process. During the charging and discharging process, the state of the battery to be tested includes: electrochemical polarization state and concentration difference polarization state, and the electrical parameters include: inherent resistance of the battery to be tested, open circuit voltage of the battery to be tested, first resistance and first capacitance of the battery to be tested in the electrochemical polarization state, and second resistance and second capacitance of the battery to be tested in the concentration difference polarization state; determining a state space model of the battery to be tested according to the electrical parameters, wherein the state space model is used to indicate the correlation between the input current of the battery to be tested, the output voltage of the battery to be tested and the electrical parameters; using a particle swarm optimization algorithm to analyze the state space model to obtain an optimal parameter value, wherein the particle swarm optimization algorithm incorporates complex order operators, and the optimal parameter value is the value of the electrical parameter that minimizes the error value, and the error value is the error value between the current measured voltage value and the current predicted voltage value of the battery to be tested at the current moment, and the current predicted voltage value is the voltage value determined according to the equivalent circuit model and the optimal parameter value.

[0078] The embodiment of the present application also provides a computer program product, including computer instructions, which implement the steps of the above method for determining the optimal parameter value when executed by a processor.

[0079] It should be noted that the various modules in the above-mentioned device for determining the optimal parameter value can be a program module (for example, a set of program instructions for implementing a certain specific function) or a hardware module. For the latter, it can be expressed in the following forms, but is not limited to this: the expression form of each of the above-mentioned modules is a processor, or the functions of each of the above-mentioned modules are implemented by a processor.

[0080] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0081] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0082] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0083] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0084] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0085] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the relevant technology or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc. Various media that can store program codes.

[0086] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for determining an optimal parameter value, characterized in that: include: Obtaining an equivalent circuit model of the battery to be tested, wherein the equivalent circuit model is used to indicate the change state of electrical parameters of the battery to be tested during the charge and discharge process. During the charge and discharge process, the state of the battery to be tested includes: an electrochemical polarization state and a concentration difference polarization state, and the electrical parameters include: an inherent resistance of the battery to be tested, an open circuit voltage of the battery to be tested, a first resistance and a first capacitance of the battery to be tested in the electrochemical polarization state, and a second resistance and a second capacitance of the battery to be tested in the concentration difference polarization state; Determining a state space model of the battery to be tested according to the electrical parameters, wherein the state space model is used to indicate a correlation relationship between an input current of the battery to be tested, an output voltage of the battery to be tested, and the electrical parameters; The state space model is analyzed using a particle swarm optimization algorithm to obtain an optimal parameter value, wherein the particle swarm optimization algorithm incorporates a complex-order operator, the optimal parameter value is the value of the electrical parameter that minimizes the error value, the error value is the error value between the current measured voltage value and the current predicted voltage value of the battery to be tested at the current moment, and the current predicted voltage value is a voltage value determined based on the equivalent circuit model and the optimal parameter value.

2. The method according to claim 1, characterized in that: Determining a state space model of the battery to be tested according to the electrical parameters includes: Determine the current value of the battery to be tested at the current moment; Determine a first voltage according to the first resistor, the first capacitor and the current value, and determine a second voltage according to the second resistor and the second capacitor, wherein the first voltage is used to indicate an electrochemical polarization effect of the battery under test during the charge and discharge process, and the second voltage is used to indicate a concentration extreme difference effect of the battery under test during the charge and discharge process; Acquire the state of charge SOC of the battery to be tested at the current moment, and determine the state variables of the state space model according to the SOC, the first voltage and the second voltage; Determine an objective function for indicating a correlation between the SOC and the open circuit voltage of the battery to be tested, and determine an output variable of the state space model according to the objective function, the first voltage, the second voltage, the current value, and the inherent resistance; The current value is determined as an input variable of the state-space model, and the state-space model is determined according to the state variable, the output variable, and the input variable.

3. The method according to claim 1, characterized in that The state space model is analyzed using a particle swarm optimization algorithm to obtain optimal parameter values, including: Acquire multiple groups of historical state information of the state space model at multiple historical moments before the current moment, and generate multiple first target particle swarms to be optimized according to the multiple groups of historical state information, wherein each of the first target particle swarms to be optimized includes multiple particles to be optimized, each group of the historical state information is a particle to be optimized, each of the historical moments corresponds to a group of the historical state information, and each group of the historical state information at least includes: the historical positions of the particles to be optimized at multiple historical moments before the current moment, the historical positions are used to indicate the values ​​of the particles to be optimized at the historical moments, and each of the particles to be optimized includes: the values ​​of the first type of parameters included in the state variables in the state space model and the state space model The output variable in the formula (a) includes the values ​​of the second type of parameters included in the output variable, wherein the first type of parameters includes: the first capacitor, the first resistor, the second capacitor and the second resistor, the values ​​of the first type of parameters include: the capacitance value of the first capacitor, the first fractional order of the first capacitor, the resistance value of the first resistor, the capacitance value of the second capacitor, the second fractional order of the second capacitor and the resistance value of the second resistor, the first fractional order is used to indicate the frequency response characteristics of the first capacitor, and the second fractional order is used to indicate the frequency response characteristics of the second capacitor, the second type of parameters includes: a plurality of coefficients in the objective function of generating the output variable, and the values ​​of the second type of parameters include: a plurality of values ​​corresponding to the plurality of coefficients; Iteratively updating the plurality of first target particle swarms to be optimized by using the particle swarm optimization algorithm until the number of iterations reaches a preset number of iterations, and outputting a target particle swarm, the target particle swarm being the result of the last iterative update; The error value corresponding to each particle in the target particle group is determined, the particle corresponding to the error value with the smallest value is determined as the target particle, and the multiple values ​​contained in the target particle are determined as the optimal parameter values.

4. The method according to claim 3, characterized in that The particle swarm optimization algorithm is used to iteratively update the plurality of first target particle swarms to be optimized, including: For each of the first target particle groups to be optimized, determining an individual optimal position and a group optimal position according to the historical state information, wherein the individual optimal position is the historical position of each particle to be optimized when the historical error value is the minimum, the historical error value is the root mean square difference between multiple predicted voltage values ​​and multiple measured voltage values ​​corresponding to multiple historical moments, and the group optimal position is determined from the multiple individual optimal positions; According to the individual optimal position and the group optimal position, the speed and position of each particle to be optimized are updated to obtain a speed update result and a position update result, wherein the speed update result is an adjustment direction and an adjustment value of the particle to be optimized when performing the iterative update, and the adjustment direction includes: increase and decrease, and the position update result is a value of the particle to be optimized after the iterative update; Determine an update method for the first target particle group to be optimized according to the speed update result and the position update result, and update the first target particle group to be optimized according to the update method, wherein the update method includes: a first update method and a second update method, wherein the first update method is used to indicate an increase in the number of particles to be optimized contained in the first target particle group to be optimized, and the second update method is used to indicate a decrease in the number of particles to be optimized contained in the first target particle group to be optimized; Determine the particle count of each of the first target particle groups to be optimized, and determine the second target particle group to be optimized for the next iterative update according to the particle count, wherein the particle count is the number of particles to be optimized contained in the first target particle group to be optimized after being updated according to the update method.

5. The method according to claim 4, characterized in that Updating the speed and position of each particle to be optimized according to the individual optimal position and the group optimal position includes: Update the speed of each particle to be optimized according to the following formula: Among them, the represents the complex order operator, the α1 represents the first fractional order, the β1 represents the second fractional order, the represents the speed update result of the particle i to be optimized after the kth iteration update, wherein k represents the number of iterations, c1 and c2 are different preset values, r1 and r2 are random numbers within the preset value range, represents the individual best position of the particle i to be optimized during the k-th iteration update, represents the position of the particle i to be optimized before the k-th iteration update, Represents the optimal position of the group when performing the k-th iterative update.

6. The method according to claim 4, characterized in that The method further comprises: updating the speed and position of each particle to be optimized according to the individual optimal position and the group optimal position; Update the position of each particle to be optimized according to the following formula: Wherein, k represents the number of iterations, represents the position update result of the particle i to be optimized after the kth iteration update. represents the position of the particle i to be optimized during the k-th iteration update, Represents the current speed of the particle i to be optimized when performing the k-th iterative update.

7. The method according to claim 4, characterized in that The updating method of the first target particle swarm to be optimized is determined according to the speed update result and the position update result, comprising: The error value corresponding to each particle to be optimized after this iterative update is determined according to the following formula: Wherein, E represents the error value, k represents the kth iteration update, M represents the total number of iteration updates, V0(k) is the measured voltage value of the battery to be tested when performing the kth iteration update, is a voltage value determined according to the equivalent circuit model, the speed update result and the position update result obtained by the k-th iteration update of the particle i to be optimized; Comparing the error value with a previous error value to obtain a comparison result, and determining the update method according to the comparison result, wherein the previous error value is an error value determined according to a previous iterative update result of the particle to be optimized; When the comparison result indicates that the error value is smaller than the previous error value, the update method is determined to be the first update method; when the comparison result indicates that the error value is greater than or equal to the previous error value, the update method is determined to be the second update method.

8. The method according to claim 4, characterized in that Updating the first target particle swarm to be optimized according to the updating method includes: updating the first target particle swarm to be optimized according to the first updating method, and updating the first target particle swarm to be optimized according to the second updating method, wherein: Updating the first target particle group to be optimized according to the first updating method includes: determining the data set consisting of the speed update result and the position update result as a new particle; adding the new particle to the particle group to be optimized, and updating the particle count; Updating the first target particle group to be optimized according to the second updating method includes: determining an error value corresponding to each particle to be optimized among a plurality of particles to be optimized currently included in the first target particle group to be optimized; determining the particle to be optimized corresponding to the error value with the largest value as the worst particle; deleting the worst particle, and updating the particle count.

9. The method according to claim 8, characterized in that Determining a second target particle group to be optimized for the next iterative update according to the particle count includes: Determine the first target particle group to be optimized whose particle count is less than a first preset threshold as a particle group to be deleted, delete the particle group to be deleted, and update the group count, wherein the first preset threshold is used to indicate the minimum number of particles to be optimized contained in each first target particle group to be optimized, and the group count is used to indicate the number of the remaining first target particle group to be optimized; When the population count is less than a second preset threshold and the number of times the second updating method is executed is zero, determining the first target particle population to be optimized to which the new particle is added as the second target particle population to be optimized; When the population count is less than the second preset threshold and the number of executions of the second update method is not zero, or when the population count is greater than or equal to the second preset threshold, the multiple first target particle populations to be optimized after this iterative update are determined as the second target particle population to be optimized.

10. A device for determining an optimal parameter value, characterized in that: include: An acquisition module, used for acquiring an equivalent circuit model of a battery to be tested, wherein the equivalent circuit model is used for indicating a change state of electrical parameters of the battery to be tested during a charge and discharge process, wherein the state of the battery to be tested during the charge and discharge process includes: an electrochemical polarization state and a concentration difference polarization state, and the electrical parameters include: an intrinsic resistance of the battery to be tested, an open circuit voltage of the battery to be tested, a first resistance and a first capacitance of the battery to be tested under the electrochemical polarization state, and a second resistance and a second capacitance of the battery to be tested under the concentration difference polarization state; A determination module, used to determine a state space model of the battery to be tested according to the electrical parameters, wherein the state space model is used to indicate a correlation relationship between an input current of the battery to be tested, an output voltage of the battery to be tested, and the electrical parameters; An analysis module is used to analyze the state space model using a particle swarm optimization algorithm to obtain an optimal parameter value, wherein the particle swarm optimization algorithm incorporates a complex-order operator, the optimal parameter value is the value of the electrical parameter that minimizes the error value, the error value is the error value between the current measured voltage value and the current predicted voltage value of the battery to be tested at the current moment, and the current predicted voltage value is a voltage value determined according to the equivalent circuit model and the optimal parameter value.

11. A non-volatile storage medium, characterized in that: The non-volatile storage medium stores a computer program, wherein the method for determining an optimal parameter value according to any one of claims 1 to 9 is executed by running the computer program on the device where the non-volatile storage medium is located.

12. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method for determining an optimal parameter value according to any one of claims 1 to 9 through the computer program.

13. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method for determining the optimal parameter value according to any one of claims 1 to 9 are implemented.

Images