Construction method of equivalent circuit model considering shelving time of single-flow zinc-nickel battery

Abstract

The invention belongs to the field of single-flow zinc-nickel battery model establishment and parameter identification, and particularly relates to a construction method of an equivalent circuit model considering the shelving time of a single-flow zinc-nickel battery, which comprises the following steps of: establishing a second-order RC equivalent circuit model; adding a resistor and a single-pole double-throw switch on the second-order RC equivalent circuit model to form a state switching equivalent circuit model; when the load current is not zero, the battery is in a working state, and a GrGWO algorithm is adopted to identify parameters of the second-order RC equivalent circuit model; and when the load current is 0, the battery is in a shelving state, the single-pole double-throw switch acts, the circuit structure is changed, and parameters of the state switching equivalent circuit model are identified by adopting an MPA algorithm. The characteristic that when the single-liquid-flow zinc-nickel battery is shelved, the time constant can change along with the increase of shelving time is considered, so that the model can better track the change of the terminal voltage of the single-liquid-flow zinc-nickel battery under a complex working condition, and the accuracy of the model is improved.

Description

Technical Field

[0001] The invention belongs to the field of single-flow zinc-nickel battery model establishment and parameter identification, and specifically is a method for constructing an equivalent circuit model taking into account the shelf time of a single-flow zinc-nickel battery. Background Art

[0002] The single-flow zinc-nickel battery is an alkaline sedimentation flow battery that uses only one electrolyte and one storage tank, and does not use an ion exchange membrane. It has the advantages of flexible site selection, simple structure, and low cost, and has a high application prospect in large-scale energy storage technology. The modeling research of flow batteries is of great significance for studying the internal mechanism of flow batteries and helping to improve the performance of flow batteries.

[0003] In the early stage of single-flow zinc-nickel battery modeling, the external characteristics of the battery were studied by establishing empirical models. Later, many researchers referred to the equivalent circuit models of other batteries for the modeling of single-flow zinc-nickel batteries and made some improvements. When establishing the model, many researchers ignored the fact that the time constant of single-flow zinc-nickel batteries will change with the increase of shelf time, which leads to the reduction of model accuracy. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an equivalent circuit model that takes into account the shelf time of a single-flow zinc-nickel battery. Taking into account that when a single-flow zinc-nickel battery is stored, the time constant will change with the increase of the storage time. This characteristic enables the model to better track the changes in the terminal voltage of the single-flow zinc-nickel battery under complex working conditions, thereby improving the accuracy of the model.

[0005] The present invention is achieved in this way.

[0006] A method for constructing an equivalent circuit model considering the shelf time of a single-flow zinc-nickel battery, comprising:

[0007] Establish a second-order RC equivalent circuit model;

[0008] Adding a resistor and a single-pole double-throw switch to the second-order RC equivalent circuit model constitutes a state switching equivalent circuit model; the second-order RC equivalent circuit model includes an open circuit voltage U connected to the input end ocv , and the open circuit voltage U ocv The connected ohmic internal resistor R0, the other end of the ohmic internal resistor R0 is connected in series with the electrochemical polarization internal resistor R1 and the concentration polarization internal resistor R2, the electrochemical polarization internal resistor R1 is connected in parallel with the electrochemical polarization capacitor C1, and the concentration polarization internal resistor R2 is connected in parallel with the concentration polarization capacitor C2; the added resistor is the resistor R3, and the resistor R3 and the concentration polarization internal resistor R2 are selected by a single-pole double-throw switch to select whether the resistor R3 or the concentration polarization internal resistor R2 is connected in parallel with the concentration polarization capacitor C2;

[0009] When the load current is not 0, the battery is in working state, and the GrGWO algorithm is used to identify the parameters of the second-order RC equivalent circuit model;

[0010] When the load current is 0, the battery is in a standby state, the single-pole double-throw switch is in action, the circuit structure changes, and the MPA algorithm is used to identify the parameters of the state switching equivalent circuit model.

[0011] Furthermore, the process of the second-order RC equivalent circuit model is: collecting data during the charge and discharge process of the single-fluid-flow zinc-nickel battery, filtering the data, determining the characteristic points of the charge and discharge curves, drawing the battery charge and discharge voltage / current change curves, and establishing a second-order RC equivalent circuit model.

[0012] Furthermore, the current-voltage relationship in the second-order RC equivalent circuit model is expressed as:

[0013]

[0014]

[0015]

[0016] Among them I R1 and I C1 are the currents flowing through the electrochemical polarization internal resistance R1 and the electrochemical polarization capacitor C1, Q1 is the charge of the electrochemical polarization capacitor C1, and I R2 and I C2 are the currents flowing through the concentration polarization internal resistance R2 and the concentration polarization capacitor C2, Q2 is the charge of the concentration polarization capacitor C2, I is the current flowing through the load, V t is the terminal voltage, and the terminal voltage in the second-order RC equivalent circuit model is expressed as:

[0017]

[0018] V t =V ocv -IR0-V1-V2

[0019] V1 is the voltage of the electrochemical polarization capacitor C1, and V2 is the voltage of the electrochemical polarization capacitor C2.

[0020] Furthermore, the GrGWO algorithm is used to identify the parameters of the second-order RC equivalent circuit model, including:

[0021] The fully charged single-flow zinc-nickel battery was subjected to pulse cycle discharge, and the characteristic curve of the battery constant current pulse discharge was drawn, and five voltage characteristic points A, B, C, D, and E were obtained, including: the single-flow zinc-nickel battery started constant current discharge at point A, stopped discharge at point C, the AB and CD segments had voltage mutations due to ohmic polarization, the BC segment battery was discharged stably, and the DE segment battery was discharged.

[0022] The average value of the voltage difference between the AB segment and the CD segment is used as the terminal voltage of the ohmic resistance for calculation. The ohmic resistance calculation formula is as follows:

[0023]

[0024] U A Indicates the voltage corresponding to point A, U B Indicates the voltage corresponding to point B, U C Indicates the voltage corresponding to point C, U D Indicates the voltage corresponding to point D;

[0025] The expression of the battery terminal voltage of the CD segment is as follows:

[0026]

[0027] The expression of the battery terminal voltage of the DE segment is as follows:

[0028]

[0029] in,

[0030] τ1=R1C1

[0031] τ2=R2C2;

[0032] The GrGWO algorithm is used to fit the zero input response curve of the expression of the battery terminal voltage in the DE segment to obtain the time constants τ1 and τ2; then the electrochemical polarization internal resistance R1 and the electrochemical polarization capacitance C1 are obtained by fitting the zero state response curve of the expression of the battery terminal voltage in the CD segment, and then the electrochemical polarization internal resistance R1 and the electrochemical polarization capacitance C1 are calculated.

[0033] Furthermore, when the battery is in a standby state, the expression of the battery terminal voltage in the DE segment is improved to:

[0034]

[0035]

[0036] A and B are the fitting coefficients required for the time constant τ2, T is the threshold. After the shelf time t exceeds T, τ2 becomes a fixed constant.

[0037] Furthermore, the parameters of the state switching equivalent circuit model identified by the MPA algorithm include: Obtain the best fit A, B, T, and obtain the time constant τ2 through the best fit A, B, T. According to R3 = τ2 / C2, obtain the resistance R3, where t is time, V m is the measured voltage value, V e It is the voltage estimation value obtained based on the improved model. According to the voltage in the idle state, it can be known that it is a function related to A, B, and T.

[0038] When the battery load has working current, a traditional second-order RC equivalent circuit model is established, and the optimized Grey Wolf Optimization Algorithm (GrGWO) is used to identify the parameters of the second-order RC equivalent circuit model. When the battery load current is 0, it is in a shelved state. An electrical component of a single-pole double-throw switch controlled by the load current is added to the second-order RC equivalent circuit model to form a state switching equivalent circuit, and the Marine Predator Algorithm (MPA) is used to identify the parameters of the added electrical components.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The equivalent circuit model established by the method adopted in the present invention has more accurate dynamic characteristics of a single-flow zinc-nickel battery, adopts an improved algorithm and adopts a more suitable algorithm under different conditions to ensure the accuracy of model parameter identification, and the proposed model will show more superior performance when the shelf time is slightly increased. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flow chart of a method for constructing an equivalent circuit model provided by an embodiment of the present invention;

[0042] Figure 2 A schematic diagram of the composition of a single-liquid-flow zinc-nickel battery provided in an embodiment of the present invention;

[0043] Figure 3 A second-order RC equivalent circuit diagram provided by an embodiment of the present invention;

[0044] Figure 4 A constant current pulse discharge curve of a common lithium-ion battery provided in an embodiment of the present invention;

[0045] Figure 5 A lithium-ion battery parameter identification flow chart provided by an embodiment of the present invention;

[0046] Figure 6 A curve showing the change of the control coefficient a with the number of iterations t provided in an embodiment of the present invention;

[0047] Figure 7 A schematic diagram of a state switching equivalent circuit model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0049] See also Figure 1 Combination Figure 3 and Figure 7 As shown, a method for constructing an equivalent circuit model considering the shelf time of a single-flow zinc-nickel battery includes:

[0050] Establish a second-order RC equivalent circuit model;

[0051] See also Figure 3 and Figure 7 As shown, a resistor and a single-pole double-throw switch are added to the second-order RC equivalent circuit model to form a state switching equivalent circuit model; the second-order RC equivalent circuit model includes an open circuit voltage U connected to the input end ocv , and the open circuit voltage U ocv The connected ohmic internal resistor R0, the other end of the ohmic internal resistor R0 is connected in series with the electrochemical polarization internal resistor R1 and the concentration polarization internal resistor R2, the electrochemical polarization internal resistor R1 is connected in parallel with the electrochemical polarization capacitor C1, and the concentration polarization internal resistor R2 is connected in parallel with the concentration polarization capacitor C2; the added resistor is the resistor R3, and the resistor R3 and the concentration polarization internal resistor R2 are selected by a single-pole double-throw switch to select whether the resistor R3 or the concentration polarization internal resistor R2 is connected in parallel with the concentration polarization capacitor C2;

[0052] When the load current is not 0, the battery is in working state, and the GrGWO algorithm (improved Grey Wolf Optimization Algorithm) is used to identify the parameters of the second-order RC equivalent circuit model;

[0053] When the load current is 0, the battery is in a standby state, the single-pole double-throw switch is in action, the circuit structure changes, and the MPA algorithm is used to identify the parameters of the state switching equivalent circuit model.

[0054] First, a new electrical component, a single-pole double-throw switch, is added to the traditional second-order RC equivalent circuit model to form a state switching equivalent circuit. Then, the solution is determined based on the load current. When the load current is 0, the battery is in a standby state, the single-pole double-throw switch is activated, the circuit structure changes, and the MPA algorithm is used to identify the model parameters. When the load current is not 0, the circuit is in a working state, and the GrGWO (improved gray wolf optimization algorithm) algorithm is used to identify the model parameters. The flow chart is as follows Figure 1 As shown:

[0055] The schematic diagram of the single-flow zinc-nickel battery is as follows Figure 2 shown.

[0056] Establish an equivalent circuit model for single-flow zinc-nickel batteries, simulate the external characteristics of single-flow zinc-nickel batteries, take single-flow zinc-nickel batteries as the research object, start from the experiment, build a charge and discharge test platform for single-flow zinc-nickel batteries, collect data during the charge and discharge process of single-flow zinc-nickel batteries, filter the data, determine the characteristic points of the charge and discharge curve, such as the starting point of rapid voltage rise and the turning point of stable rise during the charging process, draw the battery charge and discharge voltage / current change curve, and establish a second-order RC equivalent circuit model, such as Figure 3 As shown, it mainly consists of the following parts: open circuit voltage V ocv , ohmic internal resistance R0, electrochemical polarization capacitance C1, electrochemical polarization resistance R1, electrochemical polarization voltage V1, concentration polarization capacitance C2, concentration polarization resistance R2, concentration polarization voltage V2, where the open circuit voltage is the terminal voltage after the battery is disconnected and left to stand for a long time. In the identification of the parameters of the equivalent circuit model of a single-flow zinc-nickel battery, the ohmic internal resistance is used to characterize the ohmic characteristic devices composed of the electrolyte, electrode substrate, external connection circuit, etc. of the single-flow zinc-nickel battery. Due to the existence of ohmic polarization during the charge and discharge process, the overpotential caused by it will disappear instantly when the load current is zero, so it is relatively easy to identify. However, the two series RC networks are used to characterize the complex dynamic characteristics of the electrochemical polarization and concentration polarization phenomena in the battery. Electrochemical polarization is the deviation of the electrode potential caused by the sluggishness of the electrochemical reaction when the charge and discharge current changes, while concentration polarization is caused by the ion gradient in the electrolyte. The two have different time scales and are more difficult to accurately identify than ohmic internal resistance.

[0057] According to Kirchhoff's law, the current-voltage relationship in the second-order RC equivalent circuit model can be expressed as:

[0058]

[0059]

[0060]

[0061] Among them I R1 and I C1 are the currents flowing through the electrochemical polarization internal resistance R1 and the electrochemical polarization capacitor C1, Q1 is the charge of the electrochemical polarization capacitor C1, and I R2 and I C2 are the currents flowing through the concentration polarization internal resistance R2 and the concentration polarization capacitor C2, Q2 is the charge of the concentration polarization capacitor C2, I is the current flowing through the load, V t is the terminal voltage, and the terminal voltage in the second-order RC equivalent circuit model is expressed as:

[0062]

[0063] V t =V ocv -IR0-V1-V2

[0064] To establish a second-order RC circuit model of a battery, it is first necessary to select a suitable parameter identification method and estimate the model parameters, including capacitors and resistors, based on the battery discharge data and model structure. Then, the established battery model is verified through experimental verification to ensure the correctness and reliability of the model.

[0065] The fully charged lithium-ion battery is subjected to pulse cycle discharge. The characteristic curve of the battery constant current pulse discharge is as follows: Figure 4 shown.

[0066] Figure 4 A, B, C, D, and E are the five voltage characteristic points of the battery. The battery starts to discharge at a constant current at point A and stops discharging at point C. The battery in segments AB and CD undergoes voltage mutations due to ohmic polarization. The battery in segment BC discharges steadily, which can be regarded as a zero-state response. The battery in segment DE ends discharging, which can be regarded as a zero-input response. Through the data of these characteristic points, the parameters of the resistors R0, R1, R2 and the capacitors C1 and C2 in the battery can be identified. Specifically:

[0067] (1) Before point A: the battery is in a static state.

[0068] (2) Segment AB and segment CD: The battery starts discharging from a static state at point A and stops discharging at point C. The voltage at the moments corresponding to points A and C changes suddenly. This voltage change mainly occurs due to the effect of ohmic polarization. At this time, the two ends of capacitors C1 and C2 are short-circuited, and the entire circuit is only divided by the ohmic internal resistance R0 of the battery. In order to reduce experimental errors, the average value of the two voltage differences between segments AB and CD is taken as the terminal voltage of the ohmic resistance for calculation. The ohmic resistance calculation formula is as follows:

[0069]

[0070] (3) CD segment: stable discharge process, the capacitor in the circuit is charged. This process can be regarded as a zero-state response. The expression of the battery terminal voltage is as follows:

[0071]

[0072] (4) DE segment: The battery discharge is completed, but due to the presence of capacitors, the voltage response is delayed. This process can be regarded as a zero input response, and the expression of the battery terminal voltage is as follows:

[0073]

[0074] in,

[0075] τ1=R1C1

[0076] τ2=R2C2

[0077] (5) After point E: the battery is in a static state.

[0078] To identify the parameters of the second-order RC equivalent model of lithium-ion batteries based on the LS method, it is first necessary to obtain the voltage and current data of the battery through a pulse discharge experiment. The ohmic resistance of the equivalent model can be directly calculated according to the ohmic resistance calculation formula. Then, the improved Gray Wolf optimization algorithm is used to fit the zero-input response curve of equation (8) to obtain the time constants τ1 and τ2. Then, by fitting the zero-state response curve of the expression of the battery terminal voltage in the DE segment, the electrochemical polarization internal resistance R1 and the concentration polarization internal resistance R2 of the battery are obtained, and then the electrochemical polarization capacitance C1 and the concentration polarization capacitance C2 are calculated. The above process is performed for all the corresponding data of the SOC values ​​that change by 5%, and a series of battery model parameter values ​​can be obtained. The identified parameters are input into the equivalent circuit model to complete the establishment process of the second-order RC equivalent model of the lithium-ion battery. The overall flow chart is as follows. Figure 5 shown.

[0079] Among them, the improved gray wolf optimization algorithm is improved on the basis of the gray wolf optimization algorithm to identify the specific parameters of the model components. In view of the shortcomings of the GWO algorithm such as poor population diversity, slow convergence speed in the later stage, and easy to fall into local optimality, in order to improve the performance of GWO, the following improvements are proposed:

[0080] (1) Introducing the Tent chaos mapping function in the population position initialization stage

[0081] The basic GWO algorithm uses random numbers to initialize the position of the gray wolf population, which makes it difficult to retain the diversity of the population. Therefore, this paper introduces the Tent chaos mapping function (tent mapping) to replace the pseudo-random number generator of the basic GWO algorithm, making the gray wolf population initialization more uniform and random, better maintaining the diversity of the population, and improving the global search ability of the algorithm. The formula of the Tent chaos mapping function is shown as follows:

[0082]

[0083] Among them, x k 、x k+1 Represent the current and next positions respectively; q is the chaos coefficient. Within the range of q, the system is in a chaotic state. In particular, when q = 0.5, the system will have a short-period state, so the improved algorithm takes q = 0.5.

[0084] (2) Improve the control coefficient a in the process of surrounding prey

[0085] The control coefficient a of the Grey Wolf Optimization Algorithm will decrease linearly during the iteration process, resulting in the continuous weakening of the global search capability of the GWO algorithm, while the local search capability will continue to increase, which is easy to fall into the local optimum and premature maturity. In order to improve this kind of problem and improve the global search capability of the algorithm at the beginning, another update method of the control coefficient a with the number of iterations t is proposed, and its expression is shown in the formula.

[0086]

[0087] The curves of the control coefficient a before and after improvement with the number of iterations t (assuming the maximum value of t is 500) are as follows: Figure 6 As shown in the figure. The horizontal axis represents the number of iterations, and the vertical axis represents the control coefficient a. It can be seen that the control coefficient a after improvement no longer decreases linearly, but changes nonlinearly, even though it decreases from 2 to 0. In the initial stage, a is large, and the algorithm can complete the global search well. After that, a decreases rapidly, which rapidly enhances the local search ability of the algorithm and improves the convergence performance of the GWO algorithm.

[0088] At the same time, considering that the time constant of each RC parallel network will change with the increase of the shelf time when the flow battery is idle, a state switching equivalent circuit model is proposed based on the second-order RC equivalent circuit model, such as Figure 7 shown.

[0089] When the battery is in a standby state, the voltage equation of the DE segment is mainly improved. The voltage equation is as follows: since the electrochemical polarization internal resistance R1 and the electrochemical polarization capacitor C1 have not changed their structures, the time constant τ1 remains unchanged, and only the time constant τ2 is modified to become a variable related to the standby time t. Before the threshold T, it is a linear function related to the time t, and after the threshold T, it is a constant. This equation is used to identify the time constant, and the other identification processes are consistent with the conventional second-order RC equivalent circuit model acquisition method.

[0090]

[0091]

[0092] A and B are the fitting coefficients required for the time constant τ2, T is the threshold. After the shelf time t exceeds T, τ becomes a fixed constant.

[0093] The dynamic characteristics are modeled separately: when the battery load current is not zero, the state switching equivalent circuit model and the second-order RC equivalent circuit model have the same model parameters; when the battery is left idle, a new electrical component, the single-pole double-throw switch (SPDT), is introduced based on the original model to simulate the terminal voltage change of the battery when it is left idle.

[0094] Compared with the second-order RC equivalent circuit model, the state switching equivalent circuit model divides the resistance in the RC network with a larger time constant into two resistors R2 and R3 to describe the difference in the time constant in the model of the single-flow zinc-nickel battery in the shelf stage and when the load has current flowing through it. The network with a shorter time constant in the two RC networks can be considered to be approximately unchanged in the shelf stage because its transient process is relatively short. The closing mode of the single-pole double-throw switch (SPDT) is controlled by the working state of the current in the load. When there is working current in the load, the SPDT will switch to the polarization resistor R2, and the current in the same direction as the load current will flow through the RC network and charge the two polarization capacitors; when the battery is turned to the shelf state, the SPDT will switch to another polarization resistor R3. At this time, R3 and the polarization capacitor C2 form a closed loop, and the capacitor discharges through the resistor. The polarization voltage and the battery terminal voltage change with time.

[0095] The parameters of the state switching equivalent circuit model identified by the MPA algorithm include: Iterative formula Obtain the best fit A, B, T, and obtain the time constant τ2 through the best fit A, B, T. According to R3 = τ2 / C2, obtain the resistance R3, where t is time, V m is the measured voltage value, V e It is the voltage estimation value obtained based on the improved model. According to the voltage in the idle state, it can be known that it is a function related to A, B, and T.

[0096] The state switching equivalent circuit model uses the marine predator algorithm (MPA) to identify the added model parameter R3. MPA simulates the random motion of marine predators when foraging. Predators use two strategies when attacking prey, called Lévy motion and Brownian motion. The Lévy strategy is more suitable for development, while Brownian motion is more suitable for exploration. In order to balance the two strategies, the predator makes a choice based on the speed of the prey relative to the predator. MPA can be divided into three specific stages, namely initialization, elite and prey matrix development, and optimization mechanism.

[0097] Similar to other meta-heuristic algorithms, the initial solutions of this algorithm are evenly distributed in the search space of the problem. The formula for the initialization phase is as follows:

[0098] Z=LB+r(UB-LB)#(15)

[0099] Where LB and UB are the upper and lower limits of the variable, and r∈[0, 1] is a random variable.

[0100] MPA constructs two matrices, the elite matrix Elite and the prey matrix Z. The former has the number of dimensions D and the number of populations n. It consists of the predator with the best fitness function repeated n times. This matrix is ​​modified at the end of each iteration if another superior predator is identified. Z has the same dimensions as the Elite matrix, and predators update their positions based on this matrix.

[0101]

[0102] Where n represents the population size of each matrix, D represents the number of dimensions of each matrix, and Z nD represents the position of individuals in the Z matrix population, Represents the position of the individual in the Elite matrix, where 1 represents the number of iterations.

[0103] MPA is divided into three stages based on the speed ratio of prey to predator.

[0104] The first stage: Where t and t max are the number of iterations and the maximum number of iterations, respectively. This stage assumes that the prey moves in Brownian motion while the predator is constant; therefore, the prey matrix Z is updated using the following equation:

[0105]

[0106]

[0107] Where P = 0.5, R∈[0,1] is a random vector, Z i and S i represents the i-th individual in different populations, R B Represents a random number vector based on Brownian motion. Represents matrix multiplication.

[0108] The second stage, In this stage, it is assumed that the predator moves in Brownian motion and the prey moves using Levy motion. The prey matrix Z is updated in two parts. The first part Z is updated using the following formula:

[0109]

[0110]

[0111] Where R L is a random number vector based on Levy distribution, and the second part Z is updated as follows:

[0112]

[0113]

[0114] in is the parameter that adjusts the step size of the predator’s Brownian motion.

[0115] The third stage, This phase is the final stage of the search process and it assumes that the predator moves faster than the prey, so the predator moves using Levy motion while the prey remains stationary.

[0116]

[0117]

[0118]

[0119] In order to prevent the algorithm from falling into the local optimum during the iterative calculation process, an update strategy called fish aggregation device (FADs) is adopted. The FADs effect can be mathematically expressed as:

[0120]

[0121] where r∈[0, 1] represents a random vector, FAD=0.2 represents the probability of FADs affecting the optimization mechanism, U is a binary vector, and r1 and r2 are random indices of the prey matrix.

[0122] When the standby time increases slightly, the proposed equivalent circuit model will show better performance, enabling the equivalent circuit model to better track the changes in the terminal voltage of the single flow battery under complex working conditions and improve the accuracy of the model.

[0123] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing an equivalent circuit model considering the shelf time of a single-flow zinc-nickel battery, characterized in that: include: Establish a second-order RC equivalent circuit model; Adding a resistor and a single-pole double-throw switch to the second-order RC equivalent circuit model constitutes a state switching equivalent circuit model; the second-order RC equivalent circuit model includes an open circuit voltage U connected to the input end ocv , and the open circuit voltage U ocv The connected ohmic internal resistance R0, the other end of the ohmic internal resistance R0 is connected in series with the electrochemical polarization internal resistance R and the concentration polarization internal resistance R2, the electrochemical polarization internal resistance R1 is connected in parallel with the electrochemical polarization capacitor C1, and the concentration polarization internal resistance R2 is connected in parallel with the concentration polarization capacitor C2; the added resistance is the resistor R3, and the resistor R3 and the concentration polarization internal resistance R2 are selected by a single-pole double-throw switch to select whether the resistor R3 or the concentration polarization internal resistance R2 is connected in parallel with the concentration polarization capacitor C; When the load current is not 0, the battery is in working state, and the GrGWO algorithm is used to identify the parameters of the second-order RC equivalent circuit model; When the load current is 0, the battery is in a standby state, the single-pole double-throw switch is in action, the circuit structure changes, and the MPA algorithm is used to identify the parameters of the state switching equivalent circuit model.

2. The method for constructing an equivalent circuit model considering the shelf time of a single-flow zinc-nickel battery according to claim 1, characterized in that: The process of the second-order RC equivalent circuit model is: collecting data during the charge and discharge process of the single-fluid-flow zinc-nickel battery, filtering the data, determining the characteristic points of the charge and discharge curve, drawing the battery charge and discharge voltage / current change curve, and establishing the second-order RC equivalent circuit model.

3. The method for constructing an equivalent circuit model considering the shelf time of a single-flow zinc-nickel battery according to claim 2, characterized in that: The current-voltage relationship in the second-order RC equivalent circuit model is expressed as: Among them I R1 and I C1 are the currents flowing through the electrochemical polarization internal resistance R1 and the electrochemical polarization capacitor C1, Q1 is the charge of the electrochemical polarization capacitor C1, and I R2 and I C2 are the currents flowing through the concentration polarization internal resistance R2 and the concentration polarization capacitor C2, Q2 is the charge of the concentration polarization capacitor C2, I is the current flowing through the load, V t is the terminal voltage, and the terminal voltage in the second-order RC equivalent circuit model is expressed as: V t =V ocv -IR0-V1-V2 V1 is the voltage of the electrochemical polarization capacitor C1, and V2 is the voltage of the electrochemical polarization capacitor C2.

4. The method for constructing an equivalent circuit model considering the shelf time of a single-flow zinc-nickel battery according to claim 2, characterized in that: The GrGWO algorithm is used to identify the parameters of the second-order RC equivalent circuit model, including: The fully charged single-flow zinc-nickel battery was subjected to pulse cycle discharge, and the characteristic curve of the battery constant current pulse discharge was drawn, and five voltage characteristic points A, B, C, D, and E were obtained, including: the single-flow zinc-nickel battery started constant current discharge at point A, stopped discharge at point C, the AB and CD segments had voltage mutations due to ohmic polarization, the BC segment battery was discharged stably, and the DE segment battery was discharged. The average value of the voltage difference between the AB segment and the CD segment is used as the terminal voltage of the ohmic resistance for calculation. The ohmic resistance calculation formula is as follows: U A Indicates the voltage corresponding to point A, U B Indicates the voltage corresponding to point B, U c Indicates the voltage corresponding to point C, U D Indicates the voltage corresponding to point D; The expression of the battery terminal voltage of CD segment is as follows: The expression of the battery terminal voltage of segment DE is as follows: in, τ1=R1C1 τ2=R2C2; The GrGWO algorithm is used to fit the zero input response curve of the expression of the battery terminal voltage in the DE segment to obtain the time constants τ1 and τ2; then the electrochemical polarization internal resistance R1 and the electrochemical polarization capacitance C1 are obtained by fitting the zero state response curve of the expression of the battery terminal voltage in the CD segment, and then the electrochemical polarization internal resistance R1 and the electrochemical polarization capacitance C1 are calculated.

5. The method for constructing an equivalent circuit model considering the shelf time of a single-flow zinc-nickel battery according to claim 2, characterized in that: When the battery is in a standby state, the expression of the battery terminal voltage in segment DE is improved to: A and B are the fitting coefficients required for the time constant τ2, T is the threshold. After the shelf time t exceeds T, τ2 becomes a fixed constant.

6. The method for constructing an equivalent circuit model considering the shelf time of a single-flow zinc-nickel battery according to claim 5, characterized in that: The parameters of the state switching equivalent circuit model identified by the MPA algorithm include: Iterative formula Obtain the best fit A, B, T, and obtain the time constant τ2 through the best fit A, B, T. According to R3 = τ2 / C2, obtain the resistance R3, where t is time, V m is the measured voltage value, V e It is the voltage estimation value obtained based on the improved model. According to the voltage in the idle state, it can be known that it is a function related to A, B, and T.

Images