Control method of double-layer reconfigurable battery system considering path optimization
By optimizing path planning through a two-layer reconfigurable battery system architecture and a weighted directed graph model, the problems of low capacity utilization and poor SOC consistency in traditional battery packs are solved, and the efficient and safe operation of the battery system is achieved.
Patent Information
- Application Number
- CN202511456485.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-13
- Publication Date
- 2026-02-06
AI Technical Summary
Traditional battery pack designs suffer from low capacity utilization, poor SOC consistency, lack of integration of multi-dimensional parameters in path planning, and untimely fault isolation, resulting in insufficient system stability and safety.
A two-layer reconfigurable battery system architecture is adopted, and a weighted directed graph model is constructed by combining graph theory. The path planning is optimized by algorithm to achieve priority discharge of high SOC cells, minimize energy loss, and quickly isolate faulty cells, and dynamically adjust the battery connection.
It significantly improves the flexibility and reliability of battery systems, optimizes SOC consistency, reduces energy loss, and extends battery life, making it suitable for electric vehicles and energy storage power stations.
Smart Images
Figure CN121485192A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application belongs to the field of electrical engineering, and relates to a control optimization technology of a reconfigurable battery system, in particular to a control method of a double-layer reconfigurable battery system considering path optimization. BACKGROUND
[0002] With the rapid development of electric vehicles and energy storage power stations, the importance of battery management systems (BMS) in optimizing battery performance, improving system reliability, and prolonging battery life is increasingly prominent. However, traditional battery pack designs often have low capacity utilization and poor consistency, which not only leads to the inability to fully release battery energy, but also may cause overcharging or discharging of individual cells, shortening the overall service life. Therefore, how to improve the energy utilization rate of the battery pack, ensure the consistency of the state of charge (SOC) between battery cells, and effectively isolate faulty units has become a key technical challenge in battery system design.
[0003] Although existing reconfigurable battery systems have made breakthroughs in flexibility and energy management, there are still two types of deficiencies: first, the system architecture and energy scheduling mechanism have limitations. Traditional battery systems use fixed topological structures and lack flexible dynamic adjustment capabilities, making it difficult to adapt to changes in operating environments and battery health states, resulting in poor SOC consistency and low capacity utilization. At the same time, existing system energy scheduling strategies are imperfect, with insufficient response to individual SOC differences and delayed fault isolation, which can easily compromise system stability and safety. Second, the path planning algorithm has optimization deficiencies and real-time deficiencies. Existing algorithms often only focus on simple energy distribution, ignoring multi-dimensional factors such as relay losses, isolation pins, and overall energy efficiency, lacking comprehensive optimization capabilities. In complex battery systems, path planning computation is huge, making it difficult to meet real-time requirements in actual applications, thereby limiting the system's ability to quickly schedule energy and isolate faulty units during operation. SUMMARY
[0004] The present application provides a control method of a double-layer reconfigurable battery system considering path optimization, which combines the flexibility of double-layer topological structure and the advantages of multi-dimensional weight optimization, effectively improves the capacity utilization rate of the battery pack, improves the SOC consistency, significantly enhances the safety and reliability of the system, and has wide application prospects in electric vehicles and energy storage power stations.
[0005] Technical solution: To achieve the above object, the application provides a control method of a double-layer reconfigurable battery system considering path optimization, comprising the following steps:
[0006] S1: a battery mathematical model is established based on Thevenin equivalent circuit, and a double-layer reconfigurable topology structure composed of a battery cluster layer and a single battery layer is designed, and SOC and fault state are collected in real time;
[0007] S2: a weighted directed graph model is constructed based on graph theory according to the battery mathematical model and the double-layer reconfigurable topology structure;
[0008] S3: the weighted directed graph model is solved by introducing algorithm to obtain the optimal path from the source node to the sink node, so as to realize high SOC single discharge priority, energy loss minimization and rapid isolation of fault units.
[0009] Further, the expression form of the battery mathematical model in step S1 is as follows:
[0010]
[0011] In formula (1), is the output voltage; OCV is the open circuit voltage; is the battery internal resistance; is the current; is the polarization voltage, defined as formula (2):
[0012]
[0013] In formula (2), is the polarization capacitance; is the polarization voltage; the nonlinear relationship between OCV and SOC is expressed as:
[0014]
[0015] In formula (3), a i (i=0,1,2,3), b j (j=1,2) represents the coefficient of the OCV and SOC function, and SOC is defined as the ampere-hour integral form:
[0016]
[0017] In formula (4), is the total capacity, and t is the integral time.
[0018] Furthermore, in the dual-layer reconfigurable topology of step S1, the upper layer divides the battery into several battery clusters, and the lower layer realizes the access and isolation of individual batteries through independent relays.
[0019] Furthermore, in step S2, the weighted directed graph model abstracts battery clusters and individual cells as cluster nodes and individual cells, respectively, and integrates SOC consistency cost, relay loss cost and fault isolation cost into edge weights to form a multi-dimensional coupled optimization model.
[0020] Furthermore, the weighted directed graph model in step S2 is expressed as follows:
[0021] The reachability relationships between nodes are described by an adjacency matrix A, defined as follows:
[0022]
[0023]
[0024] In equation (6), and It is a weight, satisfying n is the number of switches between node i and node j; This represents the cost of energy flow through a single switch; The SOC or voltage fraction of the (i-1)th battery in the interval [-1, 1] is defined as:
[0025]
[0026] Furthermore, in step S3, by introducing The process of solving the weighted directed graph model by the algorithm includes:
[0027] A1: Initialization Algorithm, cost function;
[0028] A2: Calculate the edge weights between nodes, introduce a heuristic function, calculate the actual path and heuristic estimate, and obtain the total cost estimate;
[0029] A3: Iteratively select the node with the smallest total cost estimate, generate reachable neighbor nodes and update the cost until the sink node is found, and backtrack to obtain the optimal energy path;
[0030] A4: The BMS sends relay on / off commands based on the optimal path; individual / cluster relays within the path close, while units outside the path open; active balancing is initiated when the SOC difference exceeds the threshold; if a faulty unit is detected, its weight is immediately set to infinity. The algorithm re-finds the path, isolates faulty units, and replaces them with suboptimal units.
[0031] A5: Based on the set time interval or when the system state changes abruptly, re-collect the state information, update the graph weights, and repeat steps A1 to A4 to achieve dynamic matching between the path and the system state, ensuring optimal configuration throughout the entire lifecycle.
[0032] Furthermore, the expression for the heuristic function in step A2 is:
[0033]
[0034] In equation (8), It is the estimated total cost from the current node to the target node; This is the sum of the actual path weights from the source node to the current node n; This is a heuristic estimate of the distance from node n to the target node.
[0035] Furthermore, in the heuristic function of step A2, the actual cost of the node... The calculation formula is:
[0036] .
[0037] Furthermore, step A3 specifically includes: updating the corresponding edge weight path, processing the multidimensional edge weight conditions until the objective function weight value reaches its minimum, and outputting the optimized configuration result.
[0038] Based on the above, the innovative points of this invention can be summarized as follows:
[0039] 1. By constructing a two-layer reconfigurable topology with upper battery clusters and lower individual batteries, and configuring independent relays to achieve individual battery access and isolation control, the limitations of traditional fixed topologies are broken, and the system flexibility and fault tolerance are improved.
[0040] 2. By combining the characteristics of battery cell SOC, fault status, relay operation loss and load demand, a multi-dimensional coupled weight calculation model is established to achieve accurate matching between path planning and battery status, thereby improving capacity utilization and SOC consistency.
[0041] 3. To address the energy transfer requirements of the battery pack, utilize improvements... The algorithm optimizes the energy paths in a two-layer topology to find the optimal path scheme that balances the priority utilization of high SOC units, the minimization of relay losses, and the rapid isolation of faulty units.
[0042] 4. By using economic indicators such as extended battery cycle life, reduced energy consumption, and reduced maintenance costs, the economic costs of each part of the system can be effectively evaluated. Under the premise of ensuring stable system operation, the optimal economic efficiency of the battery pack throughout its entire life cycle can be achieved, providing technical support for the efficient utilization of battery systems in scenarios such as electric vehicles and energy storage power stations.
[0043] Beneficial effects: Compared with the prior art, the present invention has the following advantages:
[0044] 1. Adopting a dual-layer reconfigurable battery system architecture and based on The path optimization algorithm significantly improves the flexibility, efficiency, and reliability of the battery system.
[0045] 2. It can dynamically adjust the connection between individual battery cells and battery clusters, optimize SOC consistency, reduce energy loss, and achieve timely isolation of faulty cells, thereby ensuring the safe and stable operation of the battery system.
[0046] 3. The adoption of a multi-dimensional coupled weight calculation model makes energy dispatch more accurate and efficient, solving the problems of low capacity utilization and poor consistency in traditional battery management systems.
[0047] 4. It can effectively extend battery life and improve overall performance, and is widely applicable to electric vehicles, energy storage power stations and other scenarios, with significant practical value and application prospects. Attached Figure Description
[0048] Figure 1 The equivalent circuit model of Thevenin;
[0049] Figure 2 Diagram of a two-layer reconfigurable battery architecture;
[0050] Figure 3 for A diagram illustrating the algorithm's solution process for a weighted directed graph model;
[0051] Figure 4 For based on The algorithm's optimal energy path diagram for the battery pack;
[0052] Figure 5 This is the optimal energy path diagram when a faulty battery cell is isolated.
[0053] Figure 6 This is a heatmap of the weight matrix between battery cells. Detailed Implementation
[0054] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.
[0055] Example 1:
[0056] This embodiment provides a control method for a two-layer reconfigurable battery system considering path optimization, including the following steps:
[0057] S1: A mathematical model of the battery is established based on the Thevenin equivalent circuit, and a two-layer reconfigurable topology consisting of a battery cluster layer and a single cell layer is designed. At the same time, the SOC and fault status are collected in real time.
[0058] The Thevenin equivalent circuit model can effectively describe the electrical characteristics of a battery. The battery state can be approximated using an open-circuit voltage source and an equivalent internal resistance model, thus describing the battery's charging and discharging behavior under different loads and operating conditions. Figure 1 As shown, based on circuit principles, the mathematical model of a battery can be expressed as follows:
[0059]
[0060] In equation (1), Output voltage; OCV is open-circuit voltage; This refers to the battery's internal resistance. For current; The polarization voltage is defined by equation (2):
[0061]
[0062] In equation (2), Polarizing capacitor; The polarization voltage is OCV; the nonlinear relationship between OCV and SOC is expressed as:
[0063]
[0064] In equation (3), a i (i=0,1,2,3), b j (j=1,2) represents the coefficients of the OCV and SOC functions, where SOC is defined in ampere-hour integral form:
[0065]
[0066] In equation (4), Let t be the total capacity and t be the integration time.
[0067] like Figure 2 As shown, in the two-layer reconfigurable topology, the upper layer is divided into N battery clusters (each battery cluster contains 4 individual battery cells, namely cell1~cell4), and series and parallel control and fault isolation relays (equivalent to switches) are configured between the clusters. The lower layer realizes the access and isolation of individual batteries through independent relays.
[0068] S2: Based on the battery mathematical model and the two-layer reconfigurable topology, construct a weighted directed graph model based on graph theory;
[0069] The weighted directed graph model abstracts battery clusters and individual cells as cluster nodes and individual nodes respectively, and integrates SOC consistency cost, relay loss cost and fault isolation cost into edge weights to form a multi-dimensional coupled weight calculation model, namely the objective function of multi-dimensional total cost - formula (6).
[0070] The weighted directed graph model is expressed as follows:
[0071] The reachability relationships between nodes are described by an adjacency matrix A, defined as follows:
[0072]
[0073]
[0074] In equation (6), and It is a weight, satisfying n is the number of switches between node i and node j; This represents the cost of energy flow through a single switch; The SOC or voltage fraction of the (i-1)th battery in the interval [-1, 1] is defined as:
[0075]
[0076] S3: By introducing The algorithm solves the weighted directed graph model to obtain the optimal path from the source node to the sink node, achieving priority discharge of high SOC cells, minimization of energy loss, and rapid isolation of faulty cells.
[0077] The overall process of the algorithm is an ordered path search method that combines actual costs with heuristic information, aiming to efficiently find the optimal path from the starting point to the destination;
[0078] likeFigure 3 As shown, by introducing The process of solving the weighted directed graph model by the algorithm includes:
[0079] A1: Initialization Algorithm, cost function - formula (6);
[0080] A2: Calculate the edge weights a between nodes. ij A heuristic function is introduced to calculate the actual path and heuristic estimate, and the total cost estimate is obtained.
[0081] The expression for the heuristic function is:
[0082]
[0083] In equation (8), It is the estimated total cost from the current node to the target node; This is the sum of the actual path weights from the source node to the current node n; This is a heuristic estimate of the distance from node n to the target node.
[0084] The actual cost of a node The calculation formula is:
[0085]
[0086] Heuristic Cost The calculation formula can be adjusted according to the actual situation of the battery system. Generally speaking, This is an underestimation, to avoid overestimation which would reduce search efficiency.
[0087] A3: Iteratively select the node with the smallest total cost estimate, generate reachable neighbor nodes and update the cost until the sink node is found, and backtrack to obtain the optimal energy path;
[0088] The corresponding edge weight paths are updated, and the multi-dimensional edge weight conditions are processed until the objective function weight values are minimized, at which point the optimized configuration result is output. For example... Figure 4 The optimal energy path and Figure 6 The heatmap of the weight matrix between battery cells shown can all yield the optimal energy path.
[0089] A4: System Configuration and Control Execution
[0090] The BMS sends relay on / off commands based on the optimal path; individual / cluster relays within the path close, while off-path units open. Active balancing is initiated when the SOC difference exceeds a threshold. Upon detecting a faulty unit, its weight is immediately set to infinity. The algorithm re-finds the path within 10ms, isolates the faulty unit, and replaces it with a suboptimal unit, such as... Figure 5 As shown.
[0091] A5: Dynamic Optimization and Iterative Updates
[0092] Based on the interval of 0.1s or when the system state changes abruptly, the state information is re-collected, the graph weights are updated, and steps A1 to A4 are repeated to achieve dynamic matching between the path and the system state, ensuring optimal configuration throughout the entire lifecycle.
[0093] It can be seen that, based on The algorithm's two-layer reconfigurable battery system optimization configuration aims to improve capacity utilization, optimize SOC consistency, reduce energy loss, and enhance fault tolerance. It achieves optimal system configuration through a closed-loop process of model building, parameter quantization, path optimization, control execution, and dynamic updating.
[0094] Example 2:
[0095] To verify the performance of the method of this invention, a complete simulation model was built in the MATLAB simulation environment. The core parameters were configured as follows: Eight lithium-ion battery cells (numbered Node1 to Node8) were used, divided into two independent clusters according to the upper-level battery cluster architecture. Cluster 1 contains Node1 to Node4, and Cluster 2 contains Node5 to Node8. Inter-cluster relays were configured in each cluster to realize series-parallel switching. Each battery cell was configured with two access relays and one fault isolation relay, forming a complete two-layer reconfigurable topology. The initial SOC values of the battery cells were set as follows: Node1=0.7, Node2=0.9, Node3=0.3, Node4=0.6, Node5=0.8, Node6=0.75, Node7=0.42, and Node8=0.53.
[0096] Based on the above parameters, a weighted directed graph was plotted in MATLAB according to the two-layer topology. For the output load requirement containing 6 battery cells, the following was performed: The algorithm solves for the shortest path. Simulation results show that... The shortest path found by the algorithm is consistent with the theoretical analysis, accurately selecting the top six battery cells in terms of SOC (Node1→Node2→Node4→Node5→Node6→Node8). This perfectly matches the optimization objectives of prioritizing high SOC and low loss in the theoretical analysis, verifying the algorithm's ability to balance multiple objectives.
[0097] To verify the system's path replanning capability under fault scenarios, the following simulation scenario was set up: During normal discharge for 300 seconds, a short-circuit fault was triggered on Node5 using MATLAB's fault injection module, thus triggering the system's fault tolerance mechanism. The algorithm restarted the search. Since Node 5 was marked as a faulty node, all paths passing through Node 5 were excluded. During replanning, the algorithm automatically replaced it with Node 7, which has the second highest SOC, ensuring that the path still consists of high SOC cells, thus verifying the algorithm's effectiveness. The algorithm's ability to adapt and adjust under fault scenarios.
Claims
1. A control method for a two-layer reconfigurable battery system considering path optimization, characterized in that, Includes the following steps: S1: A mathematical model of the battery is established based on the Thevenin equivalent circuit, and a two-layer reconfigurable topology consisting of a battery cluster layer and a single cell layer is designed. At the same time, the SOC and fault status are collected in real time. S2: Based on the battery mathematical model and the two-layer reconfigurable topology, construct a weighted directed graph model based on graph theory; S3: By introducing The algorithm solves the weighted directed graph model to obtain the optimal path from the source node to the sink node.
2. The control method for a two-layer reconfigurable battery system considering path optimization according to claim 1, characterized in that, The mathematical model of the battery in step S1 is expressed in the following form: ； In equation (1), Output voltage; OCV is open-circuit voltage; This refers to the battery's internal resistance. For current; The polarization voltage is defined by equation (2): ； In equation (2), Polarizing capacitor; The polarization voltage is OCV; the nonlinear relationship between OCV and SOC is expressed as: ； In equation (3), a i (i=0,1,2,3), b j (j=1,2) represents the coefficients of the OCV and SOC functions, where SOC is defined in ampere-hour integral form: ； In equation (4), Let t be the total capacity and t be the integration time.
3. The control method for a two-layer reconfigurable battery system considering path optimization according to claim 2, characterized in that, In the dual-layer reconfigurable topology of step S1, the upper layer divides the battery into several battery clusters, and the lower layer realizes the access and isolation of individual batteries through independent relays.
4. The control method for a two-layer reconfigurable battery system considering path optimization according to claim 3, characterized in that, In step S2, the weighted directed graph model abstracts battery clusters and individual cells as cluster nodes and individual cells, respectively, and integrates SOC consistency cost, relay loss cost and fault isolation cost into edge weights to form a multi-dimensional coupled optimization model.
5. The control method for a two-layer reconfigurable battery system considering path optimization according to claim 4, characterized in that, The weighted directed graph model in step S2 is expressed as follows: The reachability relationships between nodes are described by an adjacency matrix A, defined as follows: ； ； In equation (6), and It is a weight, satisfying n is the number of switches between node i and node j; This represents the cost of energy flow through a single switch; The SOC or voltage fraction of the (i-1)th battery in the interval [-1, 1] is defined as: 。 6. The control method for a two-layer reconfigurable battery system considering path optimization according to claim 5, characterized in that, In step S3, by introducing... The algorithm's process for solving the weighted directed graph model includes: A1: Initialization Algorithm, cost function; A2: Calculate the edge weights between nodes, introduce a heuristic function, calculate the actual path and heuristic estimate, and obtain the total cost estimate; A3: Iteratively select the node with the smallest total cost estimate, generate reachable neighbor nodes and update the cost until the sink node is found, and backtrack to obtain the optimal energy path; A4: The BMS sends relay on / off commands based on the optimal path; individual / cluster relays within the path close, while off-path units open; active balancing is initiated when the SOC difference exceeds the threshold; upon detecting a faulty unit, its weight is immediately set to infinity. The algorithm re-finds the path, isolates faulty units, and replaces them with suboptimal units. A5: Based on the set time interval or when the system state changes abruptly, re-collect the state information, update the graph weights, and repeat steps A1 to A4 to achieve dynamic matching between the path and the system state, ensuring optimal configuration throughout the entire lifecycle.
7. The control method for a two-layer reconfigurable battery system considering path optimization according to claim 6, characterized in that, The expression for the heuristic function in step A2 is: ； In equation (8), It is the estimated total cost from the current node to the target node; This is the sum of the actual path weights from the source node to the current node n; This is a heuristic estimate of the distance from node n to the target node.
8. The control method for a two-layer reconfigurable battery system considering path optimization according to claim 7, characterized in that, In the heuristic function of step A2, the actual cost of the node The calculation formula is: 。 9. The control method for a two-layer reconfigurable battery system considering path optimization according to claim 6, characterized in that, Step A3 specifically includes: updating the corresponding edge weight paths, processing the multidimensional edge weight conditions until the objective function weight value reaches its minimum, and outputting the optimized configuration result.