Electrochemical energy storage power station battery pack simulation method capable of improving simulation speed
Through clustering analysis and n-RC equivalent circuit model, combined with the least squares algorithm of forgetting factor, partitioning and superimposing battery cell parameters, the problem of slow simulation speed of large-scale electrochemical energy storage power stations is solved, and efficient simulation is achieved Accuracy and speed balance.
Patent Information
- Application Number
- CN202410996268.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-11-12
AI Technical Summary
When the existing electrochemical energy storage power station battery pack simulation method is used to deal with large-scale energy storage power stations, the computing resources are occupied too much, resulting in slow simulation speed and it is difficult to effectively consider the inconsistency of each battery cell.
The cell cell is divided into submodule groups by cluster analysis method, and the equivalent model parameters of the series battery pack are obtained by superimposing the parameters of the submodule groups, and the parallel battery pack is solved according to the parallel relationship, combining the n-RC equivalent circuit model and forgetting The least squares algorithm of the factor performs online parameter identification.
It improves simulation speed, ensures simulation accuracy, simplifies the simulation model of large-scale electrochemical energy storage power stations, shortens calculation time, and is suitable for the verification of relay protection principles and device development of energy storage power stations.
Smart Images

Figure CN118940516A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of lithium-ion batteries for electrochemical energy storage power stations, and relates to a simulation method for a battery pack for an electrochemical energy storage power station, in particular to a simulation method for a battery pack for an electrochemical energy storage power station capable of improving the simulation speed. Background Art
[0002] In recent years, electrochemical energy storage power stations, as an important form of new energy storage, have experienced explosive growth. A large number of energy storage power stations have been built and put into operation at home and abroad on the power supply side, grid side and load side. On the power supply side, energy storage power stations can suppress the fluctuation of new energy output and improve the quality of electricity; on the grid side, energy storage power stations can meet the application needs of peak shaving and valley filling, and alleviate grid peak congestion; on the load side, energy storage power stations can participate in demand-side management and guide users to use electricity at off-peak times through time-of-use electricity prices. However, in order to meet the output voltage and current requirements, the capacity of electrochemical energy storage power stations is extremely large. Once an accident occurs, the disaster is extremely severe and the economic losses are heavy. Therefore, the safety of electrochemical energy storage power stations has attracted more and more attention from scholars at home and abroad.
[0003] Simulation can simulate the operating characteristics of electrochemical energy storage power stations and is an effective tool for studying the operating stability of electrochemical energy storage power stations and formulating response control measures and preventive measures. According to the output voltage and power requirements, the battery pack of the energy storage power station is usually composed of a certain number of battery cells connected in series and then in parallel.
[0004] At present, the simulation modeling method for battery cells is relatively mature. According to the different research mechanisms and objects, battery cell models can be divided into two categories: electrochemical models and equivalent circuit models. The electrochemical model simulates the electrochemical reaction process of the battery. Although the model has high accuracy, it needs to solve a large number of time-varying partial differential equations, and the simulation takes a long time; the equivalent circuit model simulates the external characteristics of the battery. The model establishes the relationship between the external characteristics of the battery during operation and the internal state variables of the battery cell itself. The model is simple and intuitive, and the parameters are easy to identify. It can improve the simulation speed while ensuring the simulation accuracy, so it is widely used in actual engineering.
[0005] In the simulation modeling of battery packs, since the operating temperature and remaining battery capacity of each battery cell are different during operation, the internal parameters and operating status of each battery cell are also different. In the simulation process of the battery pack of the energy storage power station, the inconsistency of each battery cell needs to be considered. However, for large-scale energy storage power stations, the number of battery cells is huge. If all battery cells are first parameterized and then calculated in series and parallel, it will take up too many computing resources and seriously affect the simulation speed. In the actual operation of the energy storage power station, some battery cells will be in similar operating states under the action of the energy management system. Based on this background, it is necessary to study a simulation method for electrochemical energy storage power stations that can improve the simulation speed.
[0006] After searching, no prior art documents identical or similar to the present invention were found. Summary of the invention
[0007] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a simulation method for an electrochemical energy storage power station battery pack that can improve the simulation speed. In view of the problem of slow simulation speed of the existing simulation technology, the differences between each battery cell are considered, and a battery grouping modeling method based on cluster analysis is proposed, which can improve the simulation speed while ensuring the simulation accuracy.
[0008] The present invention solves the practical problem by adopting the following technical solutions:
[0009] A method for simulating a battery pack of an electrochemical energy storage power station, which can improve simulation speed, comprises the following steps:
[0010] Step 1: Establish a battery cell simulation model:
[0011] Step 2: Based on the battery cell simulation model established in step 1, a cluster analysis method is used to divide each series battery group into submodule groups;
[0012] Step 3, according to the series relationship, superimpose the parameters of each submodule group divided in step 2 to obtain the equivalent model parameters of the series battery pack;
[0013] Step 4, based on the equivalent model parameters of the series battery pack obtained in step 3, solve the parallel battery pack when the series battery packs are connected in parallel.
[0014] Moreover, the specific steps of step 1 include:
[0015] Step 1-1, establishing a battery cell n-RC equivalent circuit model;
[0016] Step 1-2, establishing a mathematical expression of a battery cell n-RC equivalent circuit model;
[0017] Step 1-3, according to the requirements of simulation accuracy and simulation speed, adjust the number of RC network structures of the battery cell n-RC equivalent circuit model to determine the size of the number n of RC network structures.
[0018] Moreover, the specific method of steps 1-2 is:
[0019] According to the battery cell n-RC equivalent circuit model established in step 1-1, a mathematical relationship between the battery cell output voltage and input current based on the time domain is established, and the system equation of the model is obtained according to the bilinear transformation;
[0020] The mathematical expression construction process of the n-RC equivalent circuit model is as follows:
[0021] (1) First, according to Kirchhoff's law and Laplace transform, the relationship between the output voltage and input current of the battery cell is obtained:
[0022]
[0023] Where U t is the battery cell terminal voltage, U oc is the open circuit voltage of the battery cell, i L is the current of the battery cell, R Dn is the polarization internal resistance of the battery cell, C Dn is the polarization capacitance of the battery cell;
[0024] The transfer function of the model is obtained as:
[0025]
[0026] (2) The relationship between the output voltage and input current of the battery cell based on the time domain can be obtained from the transfer function as follows:
[0027]
[0028] According to the bilinear change law Finally, we get:
[0029]
[0030] In the formula, c i (i=1,2,......2n+1) is the coefficient to be identified of the model;
[0031] Then the data matrix Φ of the system n,k and the parameter matrix θ n,k for:
[0032]
[0033] The system equations of the equivalent circuit model can be simplified to:
[0034] y k =Φ n,k θ n,k
[0035] Among them, y k is the output voltage.
[0036] Moreover, the specific steps of step 2 include:
[0037] Step 2-1, obtaining a characteristic data set of each battery cell, wherein the obtained characteristic data set specifically includes the remaining power SOC, the current current, voltage and temperature of the battery cell;
[0038] Step 2-2, since the range and unit of each characteristic value in the data set are different, the maximum and minimum value normalization method is used to standardize the parameters of the battery cell;
[0039] The maximum and minimum normalization formula is:
[0040]
[0041] Step 2-3, on the basis of data standardization, select K battery cells as initial cluster centers, calculate the distances from the remaining battery cells to the K cluster centers, and classify each battery cell into the K clusters according to the principle of minimum distance;
[0042] The calculation formula is as follows:
[0043]
[0044] Step 2-4, after all battery cells are divided into K submodule groups, the position of the centroid of each submodule group is recalculated according to the division situation, and all battery cells are re-divided;
[0045] Step 2-5, repeat steps 2-3 and 2-4 until the centroid no longer changes or the maximum number of iterations is reached.
[0046] Moreover, the specific steps of step 3 include:
[0047] Step 3-1, denormalize the parameters of the centroid of each submodule group to obtain the original input data;
[0048] The denormalization formula is as follows:
[0049] x ij =(x max -x min )x′ ij +x min
[0050] Step 3-2, performing online parameter identification on the original input data of the centroid of each submodule group, and obtaining the equivalent model parameters of the submodule group according to the number of battery cells contained in each submodule group.
[0051] Step 3-3, based on the series relationship, the equivalent parameters of each submodule group are superimposed to obtain the equivalent model parameters of the series battery group.
[0052] Moreover, the specific method of step 3-2 is:
[0053] The least squares algorithm based on the forgetting factor is used to achieve reliable estimation of system parameters. An online parameter identification method based on FF is constructed. First, the input data variables are constructed based on the sampled battery cell current, voltage, temperature and other data. Then, the corresponding gain and covariance matrix are calculated to carry out online parameter identification and update.
[0054] The calculation process of the algorithm is as follows:
[0055] For the system equations:
[0056] y k =Φ k θ k +e Ls,k
[0057] where e Ls,k is a stationary zero-mean white noise, we have:
[0058]
[0059] The equivalent model parameters of each submodule group are shown in the following formula, taking n=1 as an example:
[0060] U ocv , k =m k U ocv , kz
[0061] R i,k =m k R i,kz
[0062] R D,k =m k R D,kz
[0063]
[0064] Among them, U ocv,k , R i,k , R D,k , C D,k Respectively represent the equivalent open circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the submodule group, m k Indicates the number of battery cells in the submodule group, U ocv,kz , R i,kz , R D,kz , C D,kz They represent the open circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the centroid of the submodule group, respectively.
[0065] Moreover, the equivalent model parameters of the series battery pack in step 3-3 are as follows:
[0066]
[0067] Among them, U ocv , R i , R D , C D They represent the equivalent open-circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the series battery pack, respectively.
[0068] Moreover, the specific method of step 4 is:
[0069] When the series battery groups are connected in parallel, the parallel equations are written and solved based on the relationship that the terminal voltages are equal and the sum of the branch currents is equal to the total current;
[0070] The parallel equation is shown below:
[0071] U s =U s,1 =U s,2 =…=U s,p
[0072] I s =I s,1 +I s,2 +…+I s,p
[0073] Among them, U s ,I s are the voltage and current of the battery string, U s,p ,I s,p are the voltage and current of the pth battery string respectively.
[0074] Advantages and beneficial effects of the present invention:
[0075] The present invention provides a method for simulating a battery pack of an electrochemical energy storage power station that can improve the simulation speed, and can be applied to the field of electrochemical energy storage power station simulation technology. The method includes: obtaining the operating parameters of each battery cell including current, voltage, operating temperature, and remaining capacity, dividing each series battery pack into multiple submodule groups by standardizing the parameters and then using a cluster analysis method, performing online parameter identification on each submodule group, and then superimposing the equivalent parameters of the submodule groups according to the series relationship to obtain an equivalent circuit model of the series battery group; writing and solving the parallel relationship equation according to the relationship that the terminal voltage of the parallel battery group is equal and the sum of the branch currents is equal to the total current. The present invention can effectively improve the simulation speed and save computing resources during the simulation process by solving the method of dividing the series battery groups into groups.
[0076] 2. Compared with offline identification of battery cell parameters, the present invention has higher simulation accuracy in online identification of battery cell parameters; the battery cell model can be flexibly adjusted according to simulation requirements;
[0077] 3. The present invention takes into account the inconsistency of each battery cell. Compared with the use of a simple equivalent model for all battery cells, the present invention can effectively improve the simulation accuracy. At the same time, the cluster analysis method is used to first divide the battery cells with similar operating states into sub-module groups. The simulation is performed using the sub-module groups as the basic simulation unit, which can effectively improve the simulation speed.
[0078] 4. The present invention simplifies the simulation model of large-scale electrochemical energy storage power stations, shortens the calculation time in actual simulation projects, and can provide a certain simulation basis for the verification of relay protection principles of energy storage power stations and the development of protection devices. BRIEF DESCRIPTION OF THE DRAWINGS
[0079] Figure 1 It is a simulation flow chart of the series connection of battery packs in the electrochemical energy storage power station of the present invention;
[0080] Figure 2 It is a schematic diagram of the n-RC equivalent circuit model of the present invention;
[0081] Figure 3 The present invention is a flow chart of the least squares online parameter identification method based on forgetting factor. DETAILED DESCRIPTION
[0082] The following further describes the embodiments of the present invention:
[0083] A simulation method for an electrochemical energy storage power station battery pack that can improve simulation speed, such as Figure 1 As shown, the following steps are included:
[0084] Step 1: Establish a battery cell simulation model:
[0085] The specific steps of step 1 include:
[0086] Step 1-1, establishing a battery cell n-RC equivalent circuit model;
[0087] According to the operating characteristics of battery cells, the n-RC model based on the internal resistance model has higher prediction accuracy than other types of battery models in terms of structural complexity and model prediction accuracy, and can accurately simulate the static and dynamic characteristics inside the battery.
[0088] The n-RC model consists of three parts: voltage source, ohmic internal resistance and RC network. Its structural diagram is shown in the figure below. Figure 2 As shown, the number of RC network structures is n.
[0089] Step 1-2, establishing a mathematical expression of a battery cell n-RC equivalent circuit model;
[0090] According to the battery cell n-RC equivalent circuit model established in step 1-1, a mathematical relationship between the battery cell output voltage and input current based on the time domain is established, and the system equation of the model is obtained according to the bilinear transformation.
[0091] The mathematical expression construction process of the n-RC equivalent circuit model is as follows:
[0092] (3) First, according to Kirchhoff's law and Laplace transform, the relationship between the output voltage and input current of the battery cell is obtained:
[0093]
[0094] Where U t is the battery cell terminal voltage, U oc is the open circuit voltage of the battery cell, i L is the current of the battery cell, R Dn is the polarization internal resistance of the battery cell, C Dn is the polarization capacitance of the battery cell;
[0095] The transfer function of the model is obtained as:
[0096]
[0097] (4) The relationship between the output voltage and input current of the battery cell based on the time domain can be obtained from the transfer function as follows:
[0098]
[0099] According to the bilinear change law Finally, we get:
[0100]
[0101] In the formula, c i (i=1,2,......2n+1) are the coefficients to be identified of the model.
[0102] Then the data matrix Φ of the system n,k and the parameter matrix θ n,k for:
[0103]
[0104] The system equations of the equivalent circuit model can be simplified to:
[0105] y k =Φ n,k θ n,k
[0106] Among them, y k is the output voltage.
[0107] Step 1-3, according to the requirements of simulation accuracy and simulation speed, adjust the number of RC network structures of the battery cell n-RC equivalent circuit model to determine the size of the number n of RC network structures;
[0108] The accuracy of the model is related to the size of n. The larger n is, the higher the model accuracy is. However, in actual engineering applications, n cannot be infinite, because the larger n is, the higher the model complexity is and the slower the simulation speed is. Therefore, in actual applications, the size of n can be flexibly selected to meet the simulation requirements.
[0109] Step 2: Based on the battery cell simulation model established in step 1, a cluster analysis method is used to divide each series battery group into submodule groups;
[0110] The specific steps of step 2 include:
[0111] Step 2-1, obtaining a characteristic data set of each battery cell, wherein the obtained characteristic data set specifically includes the remaining power SOC, the current current, voltage and temperature of the battery cell;
[0112] Step 2-2, since the range and unit of each characteristic value in the data set are different, the maximum and minimum value normalization method is used to standardize the parameters of the battery cell;
[0113] The maximum and minimum normalization formula is:
[0114]
[0115] Step 2-3, on the basis of data standardization, select K battery cells as initial cluster centers, calculate the distances from the remaining battery cells to the K cluster centers, and classify each battery cell into the K clusters according to the principle of minimum distance;
[0116] The calculation formula is as follows:
[0117]
[0118] Step 2-4, after all battery cells are divided into K submodule groups, the position of the centroid of each submodule group is recalculated according to the division situation, and all battery cells are re-divided;
[0119] Step 2-5, repeat steps 2-3 and 2-4 until the centroid no longer changes or the maximum number of iterations is reached.
[0120] Step 3, according to the series relationship, superimpose the parameters of each submodule group divided in step 2 to obtain the equivalent model parameters of the series battery pack;
[0121] The specific steps of step 3 include:
[0122] Step 3-1, denormalize the parameters of the centroid of each submodule group to obtain the original input data;
[0123] The denormalization formula is as follows:
[0124] x ij =(x max -x min )x′ ij +x min
[0125] Step 3-2, performing online parameter identification on the original input data of the mass center of each submodule group, and obtaining equivalent model parameters of the submodule group according to the number of battery cells contained in each submodule group;
[0126] The specific method of step 3-2 is:
[0127] The least squares algorithm based on the forgetting factor (FF) is used to achieve reliable estimation of system parameters. The calculation flow chart of the online parameter identification method based on FF is as follows: Figure 3 As shown, firstly, the input data variables are constructed based on the sampled battery cell current, voltage, temperature and other data, and then the corresponding gain and covariance matrix are calculated to carry out online parameter identification and update;
[0128] The calculation process of the algorithm is as follows:
[0129] For the system equations:
[0130] y k =Φ k θ k +e Ls,k
[0131] where e Ls,k is a stationary zero-mean white noise, we have:
[0132]
[0133] The equivalent model parameters of each submodule group are shown in the following formula (taking n=1 as an example):
[0134] U ocv,k =m k U ocv,kz
[0135] R i,k =m k R i,kz
[0136] R D,k =m k RD,kz
[0137]
[0138] Among them, U ocv,k , R i,k , R D,k , C D,k Respectively represent the equivalent open circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the submodule group, m k Indicates the number of battery cells in the submodule group, U ocv,kz , R i,kz , R D,kz , C D,kz They represent the open circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the centroid of the submodule group, respectively.
[0139] Step 3-3, based on the series relationship, the equivalent parameters of each submodule group are superimposed to obtain the equivalent parameters of the series battery group.
[0140] The equivalent model parameters of the series battery pack are shown as follows:
[0141]
[0142] Among them, U ocv , R i , R D , C D They represent the equivalent open-circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the series battery pack, respectively.
[0143] Step 4, based on the equivalent model parameters of the series battery pack obtained in step 3, solving the parallel battery pack when the series battery packs are connected in parallel;
[0144] The specific method of step 4 is:
[0145] When the series battery groups are connected in parallel, the parallel equations are written and solved based on the relationship that the terminal voltages are equal and the sum of the branch currents is equal to the total current;
[0146] The parallel equation is shown below:
[0147] U s =U s,1 =U s,2 =…=U s,p
[0148] I s =I s,1 +I s,2 +…+I s,p
[0149] Among them, U s ,I sare the voltage and current of the battery string, U s,p ,I s,p are the voltage and current of the ρth battery string respectively.
[0150] The working principle of the present invention is:
[0151] A method and device for simulating a battery pack of an electrochemical energy storage power station capable of improving simulation speed, comprising the following steps:
[0152] The present invention provides an electrochemical energy storage power station simulation method that can improve the simulation speed. In view of the problem of slow simulation speed of existing simulation technology, the present invention considers the differences between battery cells and provides a battery grouping modeling method based on cluster analysis, which can improve the simulation speed while ensuring the simulation accuracy. The main steps are:
[0153] Step 1: Select the battery cell simulation model:
[0154] Step 1-1, establish an n-RC equivalent circuit model. According to the operating characteristics of the battery cell, the n-RC model based on the internal resistance model has higher prediction accuracy than other types of battery models in terms of structural complexity and model prediction accuracy, and can accurately simulate the static and dynamic characteristics of the battery. The n-RC model consists of three parts: voltage source, ohmic internal resistance and RC network, where the number of RC network structures is n.
[0155] Step 1-2, establish a mathematical expression of the n-RC equivalent circuit model. According to the equivalent circuit model established in step 1-1, establish a mathematical relationship between the output voltage and input current of the battery cell based on the time domain, and obtain the system equation of the model according to the bilinear transformation.
[0156] Step 1-3, according to the needs of simulation accuracy and simulation speed, adjust the number of RC network structures and determine the size of n.
[0157] Step 2: Divide each series battery group into K submodule groups based on the cluster analysis method:
[0158] Step 2-1, obtaining a characteristic data set of each battery cell. The obtained characteristic data set specifically includes the remaining power SOC of the battery cell, the current current, voltage and temperature at the current moment;
[0159] Step 2-2, since the range and unit of each characteristic value in the data set are different, the maximum and minimum value normalization method is used to standardize the parameters of the battery cell;
[0160] Step 2-3, on the basis of data standardization, select K battery cells as initial cluster centers, calculate the distances from the remaining battery cells to the K cluster centers, and classify each battery cell into the K clusters according to the principle of minimum distance;
[0161] Step 2-4, after all battery cells are divided into K submodule groups, the position of the centroid of each submodule group is recalculated according to the division situation, and all battery cells are re-divided;
[0162] Step 2-5, repeat steps 2-3 and 2-4 until the centroid no longer changes or the maximum number of iterations is reached;
[0163] In step 2, each series-connected battery group is divided into a plurality of submodule groups based on a cluster analysis method:
[0164] Step 3: According to the series connection relationship, the parameters of each submodule group are superimposed to obtain the equivalent parameters of the series battery group:
[0165] Step 3-1, denormalize the parameters of the centroid of each submodule group to obtain the original input data;
[0166] Step 3-2, performing online parameter identification on the original input data of the mass center of each submodule group, and obtaining equivalent model parameters of the submodule group according to the number of battery cells contained in each submodule group;
[0167] Step 3-3, based on the series relationship, the equivalent parameters of each submodule group are superimposed to obtain an equivalent circuit model of the series battery pack.
[0168] The grouping in step 3 is superimposed to obtain an equivalent circuit model of the series battery pack.
[0169] Step 4, parallel battery pack solution:
[0170] Step 4-1, when the series battery groups are connected in parallel, write the parallel equation and solve it according to the relationship that the terminal voltage is equal and the sum of the branch currents is equal to the total current;
[0171] The innovation of the present invention lies in:
[0172] 1. The present invention applies a new simulation method for series battery packs - a battery pack classification and superposition method based on cluster analysis. The series battery packs are classified according to similar characteristics, and then different types of sub-module groups are superimposed to obtain an equivalent model of the series battery pack.
[0173] 2. The present invention takes into account the inconsistency of each battery cell during operation, more comprehensively reflects the actual operating conditions of the battery cells of the energy storage power station, and effectively improves the simulation accuracy.
[0174] 3. The simulation model of the battery cell in the present invention can select a suitable RC network order according to actual simulation requirements.
[0175] 3. The present invention improves the simulation speed by dividing the sub-module groups, improves the simulation accuracy by online parameter identification, effectively balances the requirements of simulation accuracy and speed, and constructs a complete electrochemical energy storage power station battery simulation method and equipment.
[0176] It should be emphasized that the embodiments of the present invention are illustrative rather than restrictive. Therefore, the present invention includes but is not limited to the embodiments described in the specific implementation modes. Any other implementation modes derived by those skilled in the art based on the technical solutions of the present invention also fall within the scope of protection of the present invention.
Claims
1. A method for simulating a battery pack of an electrochemical energy storage power station that can improve simulation speed, characterized in that: The following steps are involved: Step 1: Establish a battery cell simulation model: Step 2: Based on the battery cell simulation model established in step 1, a cluster analysis method is used to divide each series battery group into submodule groups; Step 3: According to the series relationship, the parameters of each submodule group divided in step 2 are superimposed to obtain the equivalent model parameters of the series battery pack; Step 4: Based on the equivalent model parameters of the series battery pack obtained in step 3, solve the parallel battery pack when the series battery packs are connected in parallel.
2. A method for simulating a battery pack of an electrochemical energy storage power station capable of improving simulation speed according to claim 1, characterized in that: The specific steps of step 1 include: Step 1-1, establishing a battery cell n-RC equivalent circuit model; Step 1-2, establishing a mathematical expression of a battery cell n-RC equivalent circuit model; Step 1-3, according to the requirements of simulation accuracy and simulation speed, adjust the number of RC network structures of the battery cell n-RC equivalent circuit model to determine the size of the number n of RC network structures.
3. A method for simulating an electrochemical energy storage power station battery pack capable of improving simulation speed according to claim 2, characterized in that: The specific method of steps 1-2 is: According to the battery cell n-RC equivalent circuit model established in step 1-1, a mathematical relationship between the battery cell output voltage and input current based on the time domain is established, and the system equation of the model is obtained according to the bilinear transformation; The mathematical expression construction process of the n-RC equivalent circuit model is as follows: (1) First, according to Kirchhoff's law and Laplace transform, the relationship between the output voltage and input current of the battery cell is obtained: Where U t is the battery cell terminal voltage, U oc is the open circuit voltage of the battery cell, i L is the current of the battery cell, R Dn is the polarization internal resistance of the battery cell, C Dn is the polarization capacitance of the battery cell; The transfer function of the model is obtained as: (2) The relationship between the output voltage and input current of the battery cell based on the time domain can be obtained from the transfer function as follows: According to the bilinear change law Finally, we get: In the formula, c i (i=1, 2, ... 2n+1) is the coefficient to be identified of the model; Then the data matrix Φ of the system n,k and the parameter matrix θ n,k for: The system equations of the equivalent circuit model can be simplified to: y k =Φ n,k i r,k Among them, y k is the output voltage.
4. The electrochemical energy storage power station battery pack simulation method capable of improving simulation speed according to claim 1, characterized in that: The specific steps of step 2 include: Step 2-1, obtaining a characteristic data set of each battery cell, wherein the obtained characteristic data set specifically includes the remaining power SOC, the current current, voltage and temperature of the battery cell; Step 2-2, since the range and unit of each characteristic value in the data set are different, the maximum and minimum value normalization method is used to standardize the parameters of the battery cell; The maximum and minimum normalization formula is: Step 2-3, on the basis of data standardization, select K battery cells as initial cluster centers, calculate the distances from the remaining battery cells to the K cluster centers, and classify each battery cell into the K clusters according to the principle of minimum distance; The calculation formula is as follows: Step 2-4, after all battery cells are divided into K submodule groups, the position of the centroid of each submodule group is recalculated according to the division situation, and all battery cells are re-divided; Step 2-5, repeat steps 2-3 and 2-4 until the centroid no longer changes or the maximum number of iterations is reached.
5. The electrochemical energy storage power station battery simulation method capable of improving simulation speed according to claim 1, characterized in that: The specific steps of step 3 include: Step 3-1, denormalize the parameters of the centroid of each submodule group to obtain the original input data; The denormalization formula is as follows: x ij =(x max -x min )x ij +x min Step 3-2, performing online parameter identification on the original input data of the centroid of each submodule group, and obtaining the equivalent model parameters of the submodule group according to the number of battery cells contained in each submodule group. Step 3-3, based on the series relationship, the equivalent parameters of each submodule group are superimposed to obtain the equivalent model parameters of the series battery group.
6. The electrochemical energy storage power station battery simulation method capable of improving simulation speed according to claim 5, characterized in that: The specific method of step 3-2 is: The least squares algorithm based on the forgetting factor is used to achieve reliable estimation of system parameters. An online parameter identification method based on FF is constructed. First, the input data variables are constructed based on the sampled battery cell current, voltage, temperature and other data. Then, the corresponding gain and covariance matrix are calculated to carry out online parameter identification and update. The calculation process of the algorithm is as follows: For the system equations: y k =Φ k i k +e Ls,k where e Ls,k is a stationary zero-mean white noise, we have: The equivalent model parameters of each submodule group are shown in the following formula, taking n=1 as an example: U ocv,k =m k U ocv,kz R i,k =m k R i,kz R D,k =m k R D,kz Among them, U ocv,k , R i,k , R D,f , C D,k Respectively represent the equivalent open circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the submodule group, m k Indicates the number of battery cells in the submodule group, U ocv,kz , R i,kz , R D,kz , C D,kz They represent the open circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the centroid of the submodule group, respectively.
7. The electrochemical energy storage power station battery simulation method capable of improving simulation speed according to claim 5, characterized in that: The equivalent model parameters of the series battery pack in step 3-3 are shown in the following formula: Among them, U ocv , R i , R D , C D They represent the equivalent open-circuit voltage, ohmic resistance, polarization resistance, and polarization capacitance of the series battery pack, respectively.
8. The electrochemical energy storage power station battery pack simulation method capable of improving simulation speed according to claim 1, characterized in that: The specific method of step 4 is: When the series battery groups are connected in parallel, the parallel equations are written and solved based on the relationship that the terminal voltages are equal and the sum of the branch currents is equal to the total current; The parallel equation is shown below: IN s =U s,1 =U s,2 =...=U s,p I s =I s,1 +I s,2 +…+I s,p Among them, U s ,I s are the voltage and current of the battery string, U s,p ,I s,p are the voltage and current of the pth battery string respectively.
Citation Information
Cited By
A battery system simulation method and system based on cell equivalent circuit modularization
CN122735603A
