Dynamic reconfigurable battery system energy control method and device based on state space matrix

Abstract

The invention discloses a dynamic reconfigurable battery system energy control method and device based on a state space matrix, and relates to the technical field of battery management, and the method comprises the steps: constructing a state-of-charge space matrix and a state-of-health space matrix of a battery system based on the architecture of the dynamic reconfigurable battery system, and constructing a system state space matrix; solving an optimal series equalization path and a series confidence interval; on the basis of the architecture of the dynamic reconfigurable battery system, an echelon charging and discharging strategy is adopted, and online parallel balance of the system and maximum optimization operation during a fault are realized. According to the invention, the equalization power can be set according to actual needs, equalization between the battery modules and clusters can be realized, the flexibility is high, and the equalization efficiency is high.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery management, and particularly relates to an energy control method and device for a dynamically reconfigurable battery system based on a state space matrix. Background Art

[0002] With the in-depth construction of the new power system, new energy storage technologies have developed rapidly. Battery systems such as lithium-ion batteries have the advantages of short installation cycle, high charge and discharge efficiency, and fast power response speed, and have been widely applied on a large scale. In order to meet the actual energy, power, voltage and other application requirements of energy storage power stations, a large number of battery strings with good consistency need to be connected in series and parallel to form a battery system. However, due to factors such as manufacturing process and aging degree, there are differences in the performance of newly produced batteries, and the inconsistency between batteries will become larger as the service time increases.

[0003] To address the above problems, it is necessary to study the efficient equalization and energy control methods of battery systems. Active equalization is the main research direction at present. It realizes the transfer of energy from high-energy batteries to low-energy batteries by adding inductive and capacitive energy storage elements in the circuit, and finally realizes equalization. However, the research object of active equalization is mainly the equalization of single cells to single cells; for the equalization of battery packs to single cells and the equalization between adjacent battery packs, the existing equalization topologies have problems such as long equalization time and poor flexibility of equalization paths, and there is little research on the equalization of non-adjacent battery packs in the existing technology. In the actual use process, the distribution of batteries in the system is relatively complex, and more flexible equalization and energy control methods are needed to improve the energy efficiency of the battery system. Summary of the Invention

[0004] To solve the above technical problems, the present invention adopts the following technical solutions:

[0005] An energy control method for a dynamically reconfigurable battery system based on a state space matrix, comprising:

[0006] Step 1: Based on the architecture of the dynamically reconfigurable battery system, construct the state of charge state space matrix and the health state space matrix of the battery system, and construct the system state space matrix;

[0007] Step 2: Obtain the optimal series equalization path and the series confidence interval;

[0008] Step 3: Based on the architecture of the dynamically reconfigurable battery system, adopt a cascade charge and discharge strategy to realize the online parallel equalization of the system and the maximum optimized operation during faults.

[0009] An energy control device for a dynamically reconfigurable battery system based on a state space matrix, comprising:

[0010] System state space matrix construction module: Based on the architecture of the dynamically reconfigurable battery system, construct the state of charge space matrix and the state of health space matrix of the battery system, and construct the system state space matrix;

[0011] Optimal series balancing path and series confidence interval calculation module: Calculate the optimal series balancing path and the series confidence interval;

[0012] Balancing module: Based on the architecture of the dynamically reconfigurable battery system, adopt the stepped charge and discharge strategy to achieve online parallel balancing of the system and maximum optimized operation during faults.

[0013] An electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps of the method for controlling the energy of a dynamically reconfigurable battery system based on a state space matrix.

[0014] A non-transitory computer-readable storage medium stores a computer program. When the computer program is executed by a processor, it implements the steps of the method for controlling the energy of a dynamically reconfigurable battery system based on a state space matrix.

[0015] The present invention has the following beneficial effects:

[0016] The method for controlling the energy of a dynamically reconfigurable battery system based on a state space matrix according to the present invention performs SOC series balancing on battery modules or clusters on the basis of considering the SOH of battery modules, the number of switches, and losses. The present invention considers the SOC difference of battery modules or clusters for parallel online balancing. Even if a certain battery module fails, it can ensure the safe operation of the system and minimize the discarded power. The present invention can set the balancing power according to actual needs, can achieve the balancing between battery modules or clusters, has strong flexibility and high balancing efficiency. Description of the Drawings

[0017] Figure 1 It is a schematic diagram of the structure of a dynamically reconfigurable battery system;

[0018] Figure 2 It is a schematic diagram of the energy storage power required by a dynamically reconfigurable battery system;

[0019] Figure 3 For Figure 2 It is a schematic diagram of the SOC corresponding to the provided energy storage power;

[0020] Figure 4 It is a schematic diagram of the SOC of two parts of the system without balancing;

[0021] Figure 5 It is a schematic diagram of the energy storage power of the first part and the second part when the balancing control strategy is adopted under the condition of fault-free operation of the system;

[0022] Figure 6 Schematic diagrams of the SOC of the first part and the second part corresponding to the provided energy storage power Figure 5 ;

[0023] Figure 7 Schematic diagram of a part of Figure 6 ;

[0024] Figure 8 Schematic diagram of the energy storage power of the first part of the system when a fault occurs in the second part of the system

[0025] Figure 9 Schematic diagrams of the SOC of the first part of the system corresponding to the provided energy storage power Figure 8 ;

[0026] Figure 10 Schematic diagram of the energy storage power of the two parts of the system when a cluster of faults occurs in the second part of the system

[0027] Figure 11 Schematic diagrams of the SOC of the first part and the second part corresponding to the provided energy storage power Figure 10 ;

[0028] Figure 12 Schematic diagram of the power of the two parts of the system when a module in the second part of the system fails

[0029] Figure 13 Schematic diagrams of the SOC of the first part and the second part corresponding to the provided energy storage power Figure 12 ; Detailed implementation manners

[0030] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and 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. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0031] The dynamic reconfigurable battery system energy control method based on the state space matrix of the present invention includes: Step 1, based on the architecture of the dynamic reconfigurable battery system, construct the state of charge space matrix and the health state space matrix of the battery system to obtain the system state space matrix; Step 2, obtain the optimal series path and the series confidence interval; Step 3, based on the architecture of the dynamic reconfigurable battery system, considering the SOC difference of the parallel modules / clusters, adopt the ladder charge and discharge strategy to achieve the online parallel balance of the system and the maximum optimization operation during faults. Thus, the online series and parallel balance of the dynamic reconfigurable battery system is realized.

[0032] The structure of the dynamic reconfigurable battery system of the present invention is as follows Figure 1 shown, including battery modules M11, M21,..., M74 arranged in m rows and n columns. Among them, a column of battery modules such as M11, M21,..., M71 is a cluster; each battery module includes multiple series-connected single cells, the battery modules in each column are connected in series, and a switch is connected in series on each battery module, and a switch is connected in parallel between adjacent columns of battery modules. The switch is composed of a pair of MOSFETs with opposite freewheeling directions connected in series. By controlling the on and off of each MOSFET, the control of whether the system performs charging and discharging and the charging and discharging sequence is realized. In order to meet the requirements of actual applications, the present invention starts from two aspects: series consistency analysis and parallel consistency analysis of battery modules, realizes an energy control method for a dynamic reconfigurable battery system, and improves the overall consistency of the battery system.

[0033] 1. Series consistency analysis:

[0034] Obtain the state of health SOH and state of charge SOC of each battery module through the battery management system (BMS). Based on the state of health SOH of each battery module, construct the state of health space matrix of the system, as shown in Equation (1):

[0035] (1)

[0036] Based on the state of charge SOC of each battery module, construct the state of charge space matrix of the system, as shown in Equation (2):

[0037] (2)

[0038] Based on the state of charge space matrix of the system, calculate the probability of each SOC value appearing, as shown in Equation (3):

[0039] (3)

[0040] Among them, is the number of battery modules when the SOC is a certain value.

[0041] Arrange the probabilities of each SOC value in descending order, and select the SOC value with the highest probability as the series balancing reference value , and then screen out the battery modules with the value of in each row. If there is no battery module with the value of in a certain row, then select the battery module with the value closest to :

[0042] (4)

[0043] Among them, is the number of rows of the state-of-charge state space matrix of the system, is the number of columns of the state-of-charge state space matrix of the system, is the th row and the th column SOC value of the battery module.

[0044] The SOC value of each row can be obtained through Equation (4) and the closest battery module, thereby forming a new state-of-charge state space matrix.

[0045] Based on the state-of-health state space matrix, the SOH value with the highest occurrence probability is also selected as the series balancing reference value , and then the sum and the closest battery module are obtained to form a new state-of-health state space matrix.

[0046] Multiply the new state-of-charge state space matrix by the new state-of-health state space matrix, that is, multiply the SOC of each module by the corresponding SOH, to construct the system state space matrix, as shown in Equation (5):

[0047] (5)

[0048] If there is only one non-zero value in a certain row of the system state space matrix, select the corresponding battery module for series balancing; if there are multiple non-zero values, consider the number of conduction switches and losses, and obtain the optimal series balancing path by solving Equation (6). Solve Equation (6) through an improved genetic algorithm.

[0049] (6)

[0050] Among them, is the column number of the non-zero value in the th row of the state space matrix, and there may be multiple numbers.

[0051] The optimal series balancing path can be obtained through Equation (6), and there may be multiple optimal series balancing path schemes. Select the minimum value and the maximum value from the optimal series balancing path scheme, and establish the confidence interval of series balancing as (Ideally, the SS values after multiplying the new state-of-charge state space matrix and the new state-of-health state space matrix of all battery modules are the same, such as 0.5, but due to the existence of battery inconsistency, the SS values of different battery modules are not the same). The smaller the confidence interval, the better the effect of series balancing.

[0052] The following takes a dynamically reconfigurable battery system composed of 7 series and 4 parallel battery modules as an example to illustrate the technical solution of the present invention. The structure of the dynamically reconfigurable battery system is asFigure 1 as shown

[0053] The state-of-charge space matrix constructed based on the SOC of the battery module is shown in Table 1:

[0054] Table 1

[0055] The new state-of-charge space matrix obtained by equations (2)-(3) is shown in Table 2:

[0056] Table 2

[0057] The state-of-health space matrix constructed based on the SOH of the battery module is shown in Table 3:

[0058] Table 3

[0059] The new state-of-health space matrix obtained after screening is shown in Table 4:

[0060] Table 4

[0061] Thus, the state space matrix of the system as in equation (5) is obtained:

[0062] Table 5

[0063] The optimal series balancing path is obtained by equation (6), as shown in Table 6:

[0064] Table 6

[0065] 2. Parallel consistency analysis:

[0066] Based on the dynamic reconfigurable battery system architecture, online balancing of any battery module and battery cluster can be performed. The parallel balancing method is as follows: When the SOC of the battery modules in the same row of the dynamic reconfigurable battery system is unbalanced, all the switches in that row are closed to parallelize the battery modules, and the SOC of each battery module is sorted according to equation (7), or when the SOC of each cluster is unbalanced, the SOC of each cluster is sorted according to equation (7):

[0067] (7)

[0068] When the battery system receives a charging instruction, the battery module or cluster where it is located is preferentially charged. When charged to = at that time, and The charging power is evenly shared by the two battery modules or two clusters where they are located; then and charge preferentially at the same time. When 、 = at this time, 、 and The charging power is evenly shared by the three battery modules or three clusters where they are located, and so on until all SOCs are the same. Then, all the battery modules or each cluster in the same row share the charging power jointly, so as to achieve module-level or cluster-level online equalization.

[0069] When at this time, , . Among them, is the total charging power, is the time step, is the differential power between the battery module or cluster where it is located and the battery module or cluster where it is located, is the charging power borne by the battery module or cluster where it is located, The charging power borne by the battery module or cluster where it is located is , is the capacity of each battery module or cluster.

[0070] When the battery system receives a discharge command, the battery module or cluster where it is located discharges preferentially. When discharging to = at this time, and The discharging power is evenly shared by the two battery modules or two clusters where they are located, and so on until all SOCs are the same. The battery modules in the same row or each cluster share the discharging power jointly, so as to achieve module-level or cluster-level online equalization. During the equalization process, if any battery module or cluster exceeds the upper and lower limits of the SOC, discard power will be generated.

[0071] The parallel equalization method of the battery cluster is the same as that of the battery module. Using the parallel equalization of the present invention, module-level equalization or cluster-level equalization of the battery system can be achieved.

[0072] When a certain battery module fails, the dynamic reconfigurable battery system of the present invention only needs to cut off the faulty battery module through the set control strategy, without affecting the operation of the rest. The present invention can perform real-time online equalization during operation, can ensure the system requirements to the greatest extent, reduce the power and energy matching gap, and thus achieve the maximum optimized operation when the system fails.

[0073] Setting the energy storage power required by the dynamic reconfigurable battery system is as follows Figure 2 shown, and the SOC corresponding to this energy storage power is as follows Figure 3 shown. The operating range of the system SOC is 0 - 1.

[0074] (1) Fault - free operation of the system:

[0075] Divide the dynamic reconfigurable battery system in Figure 1 into two equal parts on average. The first column and the second column are the first part, and the third column and the fourth column are the second part.

[0076] Set the initial SOC of the second part to be 0.1 higher than that of the first part, and the whole system is uniformly scheduled, that is, without an equalization strategy. The power borne by the two parts is the same, and the SOC of the two parts always differs by 0.1. As shown in Figure 4 shown, when the SOC of any part exceeds the normal operating range during operation, the system generates curtailment power, and the total curtailment power is 360.2109 MW.

[0077] Adopt an equalization control strategy. During charging, the first part bears the priority, and during discharging, the second part bears the priority. The two parts reach online equalization at 76 s, that is, the SOC is the same. The energy storage power of the two parts is as shown in Figure 5 shown, the SOC corresponding to the energy storage power of the two parts is as shown in Figure 6 shown, and the local SOC is as shown in Figure 7 shown. At this time, the cumulative curtailment power generated is 180.105 MW, and the curtailment power is greatly reduced.

[0078] Under the traditional fixed - series topology structure, some systems have cluster - level equalization and can achieve the equalization effect of Figure 5 and Figure 6 ; however, the traditional structure cannot perform module - level equalization, and the dynamic reconfigurable system can achieve the equalization effect of Figure 5 and Figure 6 according to the equalization strategy of the present invention.

[0079] (2) Faulty operation of the system:

[0080] Set the second - part system to be faulty, and only the first - part system responds to the demand, which will generate more curtailment power. The cumulative curtailment power is 2418.2972 MW. The energy storage power of the first - part system is as shown in Figure 8 shown, and the SOC of the first - part system is as shown in Figure 9 shown.

[0081] When a cluster in the second part fails, the energy storage power is distributed according to the capacity ratio, that is, the first part is allocated 2 / 3 of the energy storage power, and the second part is allocated 1 / 3 of the energy storage power. Since a cluster of batteries is removed, it cannot fully meet Figure 2The power demand in will generate curtailment power, and the cumulative curtailment power is 889.20752 MW. The energy storage power and SOC of the two parts are as Figure 10 and Figure 11 shown.

[0082] When a module in the second part fails, the energy storage power is distributed according to the capacity ratio. Due to the failure, curtailment power will be generated, and the cumulative curtailment power is 282.81944 MW. The energy storage power and SOC of the two parts are as Figure 12 and 13 shown.

[0083] Under the traditional fixed series topology, there are cluster-level fault equalizations in some systems, which can achieve Figures 8 to 11 equalization effect, but cannot achieve Figure 12 and Figure 13 module-level fault equalization and operation.

Claims

1. A dynamic reconfigurable battery system energy control method based on state space matrix, characterized in that: include: Step 1: Based on the architecture of the dynamically reconfigurable battery system, the state-of-charge space matrix and the health state space matrix of the battery system are constructed, and the system state space matrix is ​​constructed; Step 2: Obtain the optimal series equilibrium path and series confidence interval; Step 3: Based on the architecture of the dynamically reconfigurable battery system, a cascade charge and discharge strategy is adopted to achieve online parallel balancing of the system and maximum optimized operation in the event of a fault.

2. The method for controlling the energy of a dynamically reconfigurable battery system based on a state space matrix according to claim 1, characterized in that: In step 1, the dynamically reconfigurable battery system includes m rows and n columns of battery modules, the battery modules in each column are connected in series, and a switch is connected in series on each battery module, and a switch is connected in parallel between adjacent columns of battery modules. The switch is composed of a pair of MOSFETs in opposite freewheeling directions connected in series, and the system is controlled to charge and discharge and the order of charging and discharging by controlling the opening and closing of each MOSFET.

3. The method for dynamically reconfigurable battery system energy control based on state space matrix according to claim 1 is characterized in that: In step 1, the state of health SOH and state of charge SOC of each battery module of the dynamically reconfigurable battery system are obtained through the battery management system, and the system's state of health space matrix is ​​constructed based on the state of health SOH of each battery module, and the system's state of charge space matrix is ​​constructed based on the state of charge SOC of each battery module.

4. The method for controlling energy of a dynamically reconfigurable battery system based on a state space matrix according to claim 3 is characterized in that: In step 1, the probability of each SOC value occurring is calculated based on the state of charge space matrix: （3） in, is the number of battery modules when the SOC is a certain value; Arrange the probability of each SOC value in descending order, and select the SOC value with the highest probability as the series balance reference value , and then filter out each row with the value Battery module, if a row does not exist, the value is If the battery module has a value closest to Battery module: （4） in, is the number of rows of the system’s state-of-charge space matrix, is the number of columns of the system’s state-of-charge space matrix, For the system Line Column battery module SOC value; The SOC value of each row is calculated by formula (4) The closest battery module obtains a new state-of-charge space matrix; Based on the health state space matrix, the SOH value with the highest probability of occurrence is selected as the series balance reference value in the same way. , and then seek and The closest battery module forms a new health state space matrix; The new state-of-charge space matrix and the new state-of-health space matrix are multiplied by the SOC and the corresponding SOH of each module to construct the system state space matrix.

5. The method for controlling energy of a dynamically reconfigurable battery system based on a state space matrix according to claim 5, characterized in that: In step 2, if there is only one non-zero value in a row of the system state space matrix, the corresponding battery module is selected for series balancing; if there are multiple non-zero values ​​in a row of the system state space matrix, the number and loss of the on switches are considered, and the optimal series balancing path is obtained by solving equation (6): （6） in, is the state space matrix The non-zero column number of the row.

6. The method for controlling energy of a dynamically reconfigurable battery system based on a state space matrix according to claim 1, characterized in that: In step 2, select the minimum value from the optimal series balanced path solution and maximum value , the confidence interval for the tandem equilibrium is established as .

7. The method for controlling energy of a dynamically reconfigurable battery system based on a state space matrix according to claim 1, characterized in that: In step 3, online parallel balancing specifically includes: when the SOC of battery modules in the same row of the dynamically reconfigurable battery system is unbalanced, closing all switches in the row to connect the battery modules in parallel, and sorting the SOC of each battery module as shown in formula (7), or when the SOC of each cluster is unbalanced, sorting the SOC of each cluster as shown in formula (7): （7） When the battery system receives a charging command, The battery module or cluster is charged first. = hour, and The two battery modules or two clusters share the charging power; then and Priority charging, when , = hour, , and The three battery modules or three clusters share the charging power, and so on, until all SOCs are the same, all battery modules or clusters in the same row share the charging power together, thereby achieving module-level or cluster-level online balancing.

8. The method for controlling energy of a dynamically reconfigurable battery system based on a state space matrix according to claim 7, characterized in that: when hour, , ;in, is the total charging power, is the time step, for The battery module or cluster in which it is located and The difference power of the battery module or cluster, for The charging power borne by the battery module or cluster, The charging power of the battery module or cluster is , is the capacity of each battery module or cluster.

9. The method for controlling energy of a dynamically reconfigurable battery system based on a state space matrix according to claim 7, characterized in that: When the battery system receives a discharge command, The battery module or cluster is discharged first. = hour, and The two battery modules or two clusters share the discharge power evenly, and so on, until all SOCs are the same, and the battery modules or clusters in the same row share the discharge power evenly, thereby achieving module-level or cluster-level online balancing.

10. A dynamically reconfigurable battery system energy control device based on a state space matrix, characterized in that: include: System state space matrix construction module: Based on the architecture of the dynamically reconfigurable battery system, the charge state space matrix and health state space matrix of the battery system are constructed, and the system state space matrix is ​​constructed; Module for obtaining the optimal series equilibrium path and series confidence interval: obtaining the optimal series equilibrium path and series confidence interval; Balancing module: Based on the architecture of the dynamically reconfigurable battery system, a cascade charge and discharge strategy is adopted to achieve online parallel balancing of the system and maximum optimized operation in the event of a fault.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the dynamic reconfigurable battery system energy control method based on the state space matrix as described in any one of claims 1 to 9 are implemented.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the dynamic reconfigurable battery system energy control method based on the state space matrix as described in any one of claims 1 to 9 are implemented.

Images