Lithium-sodium hybrid energy storage system based on dynamic reconfigurable technology and network construction type control method

Abstract

The invention provides a lithium-sodium hybrid energy storage system based on a dynamic reconfigurable technology and a network construction type control method, and belongs to the technical field of energy storage systems and control of the energy storage systems. Through an adaptive energy storage mapping algorithm, real-time dynamic sensing and unified mapping of multi-dimensional key parameters of the lithium battery pack and the sodium battery pack are realized, and efficient cooperation of heterogeneous energy storage units is guaranteed. A variable topology energy storage interface control algorithm is adopted, a lithium-sodium interface structure is dynamically adjusted according to a system state and a scheduling demand, and optimal distribution of energy flow is realized. A multi-dimensional evolution performance optimization algorithm is introduced, and full-time-domain adaptive optimization and intelligent scheduling are carried out on indexes such as efficiency, response speed, aging rate and economical efficiency of an energy storage unit. The comprehensive performance and economical efficiency of the energy storage system are remarkably improved, the service life is prolonged, and wide application value is achieved.

Description

Technical Field

[0001] This invention belongs to the field of energy storage systems and their control technology, and more specifically relates to a lithium-sodium hybrid energy storage system based on dynamic reconfigurable technology and a grid-type control method. Background Technology

[0002] With the large-scale integration of renewable energy and the evolution of new power system structures, energy storage technology plays an increasingly important role in grid peak and frequency regulation, distributed energy management, emergency backup, and peak shaving and valley filling. Currently, lithium-ion batteries are widely used in energy storage systems due to their high energy density, high efficiency, and long lifespan, while sodium-ion batteries, with their abundant raw materials, low cost, and excellent low-temperature performance, are gradually becoming an important choice for next-generation energy storage. However, single-type energy storage units have limitations in terms of capacity utilization, response speed, aging rate, and economics, making it difficult to fully meet the needs of complex and ever-changing power application scenarios. Lithium-sodium hybrid energy storage systems, as an innovative solution combining the advantages of multiple energy storage technologies, achieve energy efficiency improvement and cost optimization through the coordinated operation of multiple energy storage components, demonstrating promising application prospects.

[0003] However, due to significant differences in performance characteristics, dynamic response, and aging mechanisms between lithium-ion and sodium-ion batteries, unified modeling, collaborative optimization, and real-time scheduling of heterogeneous energy storage systems have become critical technical challenges that urgently need to be addressed. Existing hybrid energy storage control methods primarily rely on static configuration and simplified models, which struggle to effectively address dynamic changes in energy storage unit states, nonlinear aging, and multi-objective functional requirements, resulting in suboptimal overall system performance. The development of dynamic reconfigurable technology and intelligent evolutionary optimization methods offers new insights into the deep coupling and efficient operation of heterogeneous energy storage systems. Through multi-parameter, multi-dimensional real-time sensing, intelligent interface topology adjustment, and evolutionary decision optimization, it is expected to achieve adaptive management and lifecycle performance improvement for energy storage systems. Therefore, developing control methods for hybrid energy storage systems that are compatible with both lithium and sodium batteries and possess dynamic reconfigurability and intelligent optimization capabilities has become a research hotspot and industry demand in the fields of smart grids and new energy storage. Summary of the Invention

[0004] This invention aims to address the challenges in existing lithium-sodium hybrid energy storage systems, where the significant performance differences and complex dynamic changes between lithium and sodium batteries make it difficult to achieve adaptive modeling, deep collaborative operation, and real-time intelligent scheduling of energy storage units. It overcomes technical bottlenecks such as insufficient system energy efficiency, uneven aging, delayed scheduling response, and difficulty in comprehensive performance optimization, thereby realizing efficient dynamic collaboration and intelligent optimization management throughout the entire lifecycle of heterogeneous energy storage systems.

[0005] To achieve the above objectives, the present invention employs the following technical solution: the method comprises: The adaptive dynamic mapping of the lithium-sodium dual energy storage unit enables multi-dimensional dynamic sensing of key parameters such as real-time capacity, health status, and operating temperature of the lithium battery pack and sodium battery pack, and maps the characteristic parameters of the lithium-sodium battery to the same energy dispatch reference space. The heterogeneous energy storage unit is deeply coupled and dynamically reconfigured. According to the scheduling requirements, the interface circuit structure of the lithium-sodium battery is dynamically adjusted. The coupling strength factor is introduced. Based on energy flow, response speed and aging state, the connection mode (series, parallel, hybrid) and coupling degree of the lithium-sodium unit are determined in real time. Unified virtual synchronization control for energy storage enables multiple energy storage units to coordinate through virtual synchronization machine logic in both grid-connected and islanded operation scenarios, achieving frequency, voltage support, and inertia regulation. Based on behavioral intelligence, dynamic energy allocation and fault self-healing are achieved by learning from the historical behavioral data and abnormal patterns of energy storage units. This enables forward scheduling of energy allocation and local fault self-healing of the energy storage system. By introducing a behavioral tag library self-healing priority tree, the system can quickly identify and adjust resource scheduling strategies for atypical faults, thus maintaining stable system operation. The system performance monitoring and evolutionary adaptive optimization performs multi-dimensional real-time monitoring of the operating indicators of all system units, automatically generates the optimal combination of operating parameters based on historical evolution trends, introduces an evolutionary selection mechanism, provides parameter fine-tuning suggestions for abnormal operating conditions, and actively intervenes in scheduling strategies.

[0006] In one scheme, the adaptive dynamic mapping of the lithium-sodium dual energy storage unit includes: multi-dimensional real-time acquisition of core state parameters such as real-time capacity, health status, operating temperature, internal resistance, output voltage, and charge / discharge rate of the lithium battery pack and sodium battery pack, and normalization processing of all parameters. An adaptive weight tensor is dynamically generated based on the current reliability and priority of each energy storage unit. Through a multivariate tensor space mapping method, the normalized state parameters of the two heterogeneous energy storage units, lithium battery and sodium battery, are fused with the adaptive weights to form a unified mapping reference space vector, thereby realizing full-state adaptive mapping of different types of batteries in the energy storage system.

[0007] In one scheme, the deep coupling dynamic reconfiguration of the heterogeneous energy storage unit includes: using the unified mapping vector output by the Adaptive Energy Storage Mapping Algorithm (ASMA) as the global state basis, and dynamically calculating the comprehensive coupling strength factor based on energy flow capacity, response speed and aging state parameters; Based on the real-time changes of this factor, the reconfiguration switch matrix is ​​controlled to automatically switch the series, parallel, or hybrid interface topology of lithium battery and sodium battery energy storage units. This enables deep coupling and dynamic reconfiguration of lithium-sodium heterogeneous energy storage units, effectively improving the energy efficiency, safety, and system stability of hybrid energy storage systems under various operating conditions.

[0008] In one scheme, the unified virtual synchronous control of energy storage includes: real-time acquisition of multiple attribute state parameters such as energy surplus, output power, health status, and remaining lifespan of each energy storage unit in the lithium-sodium hybrid energy storage system, forming a decision input vector, and calculating the comprehensive response capability index of each energy storage unit through a multi-attribute weighted decision function; The maximum response capability optimization method is adopted to automatically select the dominant response energy storage unit to participate in frequency and inertia support. Based on this, the main control parameters of the virtual synchronous machine, including virtual inertia and damping, are dynamically adjusted to achieve adaptive dynamic adjustment of frequency, voltage and inertia in multiple scenarios. When the system state changes or is abnormal, the response unit and weight are automatically switched to ensure the continuous and reliable operation of the system.

[0009] In one scheme, the dynamic energy allocation and fault self-healing based on behavioral intelligence includes: establishing a behavioral tag library for each energy storage unit, combining historical operating condition tags, load response tags, charging and discharging behavior tags and abnormal alarm tags, extracting behavioral feature vectors through a self-learning mechanism, and dynamically identifying operating modes and abnormal features; By utilizing a behavioral intelligence weighted inference model, behavioral-driven weighted allocation factors are calculated in real time based on behavioral labels and behavioral feature vectors to guide energy allocation optimization and achieve priority allocation of units with high reliability, low anomaly rate, and rapid response capability. Meanwhile, based on the self-healing priority tree, combined with the unit health status and historical anomalies, the system intelligently selects and switches the optimal self-healing measures, dynamically adjusts the allocation factor and output power, and realizes resource redundancy scheduling and fault self-healing of the energy storage system.

[0010] In one scheme, the system performance monitoring and evolution adaptive optimization includes: real-time monitoring and analysis of efficiency, response speed, aging rate, and economic multi-dimensional performance indicators of each unit of the energy storage system; establishing a set of operating indicators and historical performance trajectories; and using a standardized comprehensive evaluation function to quantify the multi-dimensional performance. Evolutionary optimization of operating parameters is achieved through parameter population and genetic operators to maximize the overall system performance; based on performance trajectory difference and adaptive sensitivity threshold, abnormal operating conditions are automatically detected and parameter fine-tuning suggestions are generated and pushed to the scheduling center to achieve proactive intervention and optimization adjustment.

[0011] In one approach, the coupling strength factor is calculated based on parameters such as the energy flow rate of the energy storage unit, current load demand, and aging level to optimize the energy distribution ratio between lithium and sodium.

[0012] Beneficial effects of this invention: This invention, by introducing dynamic reconfigurable technology and intelligent optimization methods, achieves multi-dimensional collaborative management and control of different types of battery cells in a lithium-sodium hybrid energy storage system, effectively overcoming the inherent contradictions in energy efficiency, response speed, and aging management of heterogeneous energy storage units. Based on real-time state perception and data-driven modeling, the system can adaptively identify the operating characteristics and health status of each energy storage unit, and dynamically optimize energy allocation and scheduling decisions in conjunction with a grid-based control strategy, significantly improving the overall operating efficiency and reliability of the energy storage system.

[0013] Meanwhile, through flexible system topology reconfiguration and intelligent evolution algorithms, this invention achieves optimal configuration and cycle management of energy storage resources in response to changing energy demands and complex operating scenarios, effectively extending the service life of energy storage units and reducing system operation and maintenance costs. Furthermore, the solution of this invention possesses high compatibility and scalability, enabling it to adapt to the widespread application of various types and scales of energy storage systems in smart grids, distributed energy resources, and new power systems, demonstrating significant value for technology promotion and economic application. Attached Figure Description

[0014] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0015] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Typical embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0016] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. To facilitate understanding, the invention will now be described more fully with reference to the accompanying drawings. Typical embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to make the disclosure of the invention more thorough and complete.

[0017] like Figure 1 As shown, the lithium-sodium hybrid energy storage system and its grid-type control method based on dynamic reconfigurable technology specifically include: Step 1: Adaptive Dynamic Mapping of Lithium-Sodium Dual Energy Storage Units The Adaptive Energy Storage Mapping Algorithm (ASMA) is used to dynamically sense key parameters such as real-time capacity, health status, and operating temperature of lithium-ion and sodium-ion battery packs. Through an innovative multivariate tensor space mapping function, the characteristic parameters of lithium-ion and sodium-ion batteries are mapped to the same energy dispatch reference space, providing a unified basis for subsequent dynamic reconfiguration and energy allocation.

[0018] The implementation of the Adaptive Energy Storage Mapping (ASMA) algorithm first relies on multi-dimensional real-time acquisition of core state parameters of lithium-ion and sodium-ion battery packs, including but not limited to real-time capacity (C), state of health (SOH), operating temperature (T), internal resistance (R), output voltage (V), and charge / discharge rate (α). To achieve unified energy dispatching for heterogeneous energy storage units, ASMA introduces multivariate tensor space mapping.

[0019] Specifically, the combination of state parameters of each energy storage unit at time t is represented as a state tensor. Where i represents the unit category (lithium or sodium). During the mapping process, normalization is first performed for different physical quantities, using a max-min mapping: To uniformly characterize the states of lithium and sodium energy storage units within the scheduling reference space, an adaptive weight tensor is introduced. The weighting factor is self-adjusted based on the current reliability and priority of the energy storage unit, and is defined as follows: ,in This can be a mapping model extracted from experience or machine learning. Ultimately, the representation vector of the mapping reference space is... in This represents the Hadamard product of vectors. Through this multivariate tensor fusion and dynamic weighting, the ASMA algorithm achieves full-state adaptive mapping of heterogeneous lithium-sodium energy storage units, seamlessly supporting subsequent dynamic reconfiguration and energy allocation mechanisms.

[0020] Step 2: Deep Coupling and Dynamic Reconfiguration of Heterogeneous Energy Storage Units A Variable Topology Energy Storage Interface Control (VTCA) algorithm is proposed to dynamically adjust the lithium-sodium battery interface circuit structure according to the scheduling requirements of ASMA output. VTCA introduces a coupling strength factor, which determines the connection mode (series, parallel, hybrid) and coupling degree of lithium-sodium cells in real time based on energy flow, response speed, and aging state, thereby forming the optimal energy output channel.

[0021] The implementation process of the Deeply Coupled Dynamically Reconfigurable Interface (VTCA) algorithm for heterogeneous energy storage units is based on the unified mapping vector output by the ASMA algorithm in the previous stage. Based on this, the system first collects various status information of the lithium battery and sodium battery in real time, through... Obtain the global state of the current hybrid energy storage system. VTCA innovatively introduces a coupling strength factor. This factor is used to quantify the synergistic capability between lithium-sodium energy storage units under current operating conditions.

[0022] Specifically, It is determined by three types of parameters: first, the energy flow capability, which is reflected in the maximum available power of each unit. Compared with current scheduling power demand The ratio; secondly, the response speed, measured by the instantaneous power jump response coefficient of each unit. The third is the aging status, reflected in the unit's current health level. Compared with historical aging rate Taking all the above factors into consideration, the coupling strength factor is defined as: in, For the algorithm to adapt the weight coefficients, To prevent small positive numbers from being divided by zero. According to... The dynamic value of VTCA allows for switching the interface topology of lithium and sodium energy storage units by controlling the reconfigurable switching matrix, supporting series, parallel, and hybrid modes.

[0023] The specific mode selection logic can be expressed as: when Higher than the set threshold When the voltage level is high, a series connection is preferred to increase the voltage level; when the voltage level is in the intermediate range, a mixed connection is selected to balance power and redundancy; if If the current level is low, the system switches to parallel mode to enhance redundancy and security. The dynamic topology switching process is coordinated with ASMA. The output optimizes energy output across all time domains and conditions. The lithium-sodium heterogeneous energy storage unit enables deep coupling and dynamic reconfiguration, ensuring efficient and stable operation of the entire system under different operating conditions.

[0024] Step 3: Unified Virtual Synchronization Control for Energy Storage in Grid-Based Applications The Multi-Attribute State Cooperative Virtual Synchronization Algorithm (MASCVSA) enables multiple energy storage units to coordinate through virtual synchronizer logic in grid-connected / islanded operation scenarios, achieving frequency, voltage support, and inertia regulation. MASCVSA employs a multi-attribute weighted decision-making mechanism, considering not only energy and power but also factors such as health and remaining lifetime, adaptively switching the dominant response unit according to the scenario.

[0025] The core of implementing the Unified Virtual Synchronous Control (MASCVSA) algorithm for grid-connected energy storage applications lies in the deep integration of the multi-attribute states of the lithium-sodium hybrid energy storage system with the Virtual Synchronous Machine (VSG) model, enabling coordinated frequency, voltage, and inertia control in both grid-connected and islanded modes. First, the system acquires the multi-attribute state parameters of each energy storage unit in real time, including energy margin. Output power Health Remaining lifespan These constitute the decision input vector. MASCVSA designs a multi-attribute weighted decision function for complex scenarios and defines the comprehensive response capability index of the energy storage unit. )as follows: in, to The weights are adaptive and adjusted by the algorithm based on the scenario and historical operating status. This is based on real-time calculations for each energy storage unit. The collaborative decision-making module automatically selects the dominant response energy storage unit using the maximum response capability optimization method, that is, it makes... The selected units are given priority in frequency and inertia support. Subsequently, the master control parameters of the virtual synchronizer are dynamically updated based on the selected units: adjusting the virtual inertia... With damping for in, and This represents the basic inertia and damping coefficient of this unit. Under this mechanism, the energy storage system can achieve adaptive grid inertia support and voltage / frequency regulation in both grid-connected and islanded states based on multi-dimensional states such as energy, health, and lifespan.

[0026] When a sudden change occurs in the system state or a local unit malfunctions, MASCVSA coordinates the switching of response units and weights to ensure the continuity and reliability of system operation. Through MASCVSA, hybrid energy storage clusters can dynamically allocate inertia, voltage, and frequency support capabilities in a globally optimal manner, meeting the intelligent management and control needs of multiple scenarios, including grid construction and islanding.

[0027] Step 4: Dynamic energy allocation and fault self-healing based on behavioral intelligence A Behavior-Driven Energy Allocation Self-Healing Algorithm (BASH) is proposed. By self-learning from historical behavior data and abnormal patterns of energy storage units, it achieves forward-looking scheduling of energy allocation and self-healing of local faults in energy storage systems. The BASH algorithm introduces a behavior tag library self-healing priority tree, which can quickly identify and adjust resource scheduling strategies for atypical faults, maintaining stable system operation.

[0028] The core of the Behavior-Driven Energy Allocation Self-Healing Algorithm (BASH) lies in utilizing the historical behavior trajectories and real-time operational data of energy storage units, combined with a self-learning strategy to identify behavioral patterns and abnormal characteristics, thereby dynamically achieving energy allocation optimization and local fault self-healing. In the algorithm implementation process, the system first establishes a behavior tag library for each energy storage unit. Its content includes historical operating condition tags, load response tags, charging and discharging behavior tags, and abnormal alarm tags, which are formalized as follows: Each of them It describes state patterns and response events within a specific time window. As it accumulates data during operation, BASH employs a self-learning mechanism (based on clustering-based anomaly detection and pattern mining) to construct behavior vectors in a multi-dimensional feature space. ,in This represents the m-th behavioral feature of the description unit i at time t (such as SOC fluctuation amplitude, charge / discharge cycle, response delay, number of abnormal triggers, etc.).

[0029] For the energy allocation process, BASH defines a behavior-driven weighted allocation factor. It is calculated jointly by historical behavior labels and current behavior vectors: in The behavioral intelligence weight inference model comprehensively considers the characteristics of units such as high reliability, low anomaly rate, and fast response capability. The overall goal of energy allocation is to optimize global performance and system robustness, and the specific allocation decision is as follows: in To allocate the optimal output power to unit i, N represents the total power demand of the system, and N represents the total number of energy storage units.

[0030] For fault self-healing, BASH designs a self-healing priority tree. The nodes represent different self-healing measures (such as rescheduling, de-rating operation, offline maintenance, bypass switching, etc.), and their priorities are dynamically assigned based on the unit's health status, historical anomaly records, and behavioral confidence. During anomaly detection, if... and The algorithm closely matches the known atypical fault characteristics and follows the path described above. It iterates from top to bottom, automatically switching to the optimal self-healing measure under the current constraints, and adjusts in real time. and This enables redundant resource scheduling and priority support for healthy units, maintaining continuous and stable system operation. Through BASH, the energy storage system can achieve forward-looking energy allocation and efficient fault self-healing based on its own behavioral history and intelligent reasoning, ensuring the safety and resilience of complex power grids.

[0031] Step 5: System performance monitoring and adaptive evolution optimization The system performs multi-dimensional real-time monitoring of the operational metrics (efficiency, response speed, aging rate, economy, etc.) of all units and automatically generates the optimal combination of operating parameters based on historical evolution trends. An evolutionary selection mechanism is introduced to provide parameter fine-tuning suggestions for abnormal operating conditions and proactively intervene in the scheduling strategy.

[0032] The system continuously monitors and analyzes the full-time operational status of each unit in the energy storage system under multiple dimensions, including efficiency, response speed, aging rate, and economic efficiency. Real-time adaptive parameter adjustment is achieved through an evolutionary optimization framework. The system first establishes a set of operational indicators for each energy storage unit. in For efficiency, For response speed, For aging rate, For economic efficiency, the indicators can be expanded according to specific needs. MEPO dynamically extracts historical evolution data through a data sliding window mechanism to form a multi-dimensional performance trajectory stack. In the optimization decision-making process, the algorithm uses a standardized comprehensive evaluation function. The multidimensional performance of unit i at time t is measured, where Assign weights to each performance dimension. For the k-th indicator, This is the maximum value of the indicator within the system (normalized). This represents the total number of performance indicators. MEPO employs an evolutionary selection mechanism to define the parameter population. Each This represents a set of operating parameters (such as charge / discharge thresholds, current limiting settings, scheduling priorities, etc.). The algorithm compares historical evolution trends with current performance, and through genetic operators (selection, crossover, mutation), the optimization objective is to maximize the overall system performance. For detected abnormal operating conditions, MEPO uses performance trajectory differential analysis. and adaptive sensitivity threshold ,determination Automatically generate new parameter fine-tuning suggestions The data is then pushed to the scheduling center to proactively intervene in the current scheduling strategy. Through MEPO, the system can achieve proactive evolutionary optimization from global to local levels, continuously improving multi-objective performance, delaying aging, and significantly enhancing operational safety and economy.

[0033] Example: 1. System composition and experimental conditions This embodiment is based on a hybrid energy storage system with a total capacity of 100kWh, comprising a 60kWh lithium battery pack and a 40kWh sodium battery pack. Each battery pack is equipped with an independent BMS, and the system employs an Adaptive Storage Mapping (ASMA) algorithm, a Variable Topology Coupling Interface Control (VTCA) algorithm, and a Multidimensional Evolutionary Performance Optimization (MEPO) algorithm for coordinated scheduling. The experimental scenario simulates the energy storage system serving the formation operation of a microgrid, with deep coupling and self-reconfiguration under 24-hour dynamic load and ambient temperature changes.

[0034] 2. Energy Storage Unit Core Parameter Acquisition and Initial Mapping (ASMA) At the start of the experiment, key parameters of the lithium and sodium batteries were collected in real time, and the relevant data are shown in Table 1. Table 1: Key state parameters of the energy storage unit (time t0) Each parameter is normalized using a maximum-minimum method, mapping them all to the [0,1] interval. For example, SOH is normalized to 0.978 (lithium) and 0.932 (sodium).

[0035] An adaptive weight tensor is introduced: In this stage, lithium batteries, due to their larger capacity and better state of equilibrium (SOH), have an initial priority weight of 0.65, while sodium batteries have a weight of 0.35. Based on the mapping function, the system obtains a unified scheduling reference representation: 3. Coupled Dynamic Reconfiguration and Energy Flow Control (VTCA) The system adjusts the lithium-sodium interface topology in real time based on the fusion status and scheduling requirements output by ASMA. During the peak load period from 8 to 12 o'clock, the system detects an increase in lithium battery temperature (T rises to 34.2℃) and an increased aging rate. At the same time, the sodium battery has high remaining capacity, so it adjusts to the "parallel-shift optimization" mode: increasing the instantaneous discharge ratio of the sodium battery to 70%.

[0036] Table 2: Interface Dynamic Reconfiguration Control Data (Time t3) The solution effectively reduces the temperature rise and SOOH (State of Health) rate of lithium batteries.

[0037] 4. Multidimensional Evolutionary Performance Optimization (MEPO) Process The system continuously records multidimensional performance indicators for each unit and extracts the average over one hour using a data sliding window. Some statistics are shown in Table 3: Table 3: Multidimensional performance statistics and normalization (10:00-11:00) The weights for each indicator were set as follows: efficiency (0.3), response rate (0.25), aging rate (0.25), and economy (0.2). MEPO detected that the lithium battery was aging too quickly, automatically lowering its discharge threshold while simultaneously increasing the priority of the sodium battery, optimizing parameters, and pushing new operating conditions. The system employs genetic evolution, actively evolving to the "optimized 2" parameters within 10 minutes, significantly improving system safety and economy.

[0038] 5. Comparative Analysis of Operational Results After 24 hours of continuous operation, the lithium-sodium system performed as follows: Table 4: Comparison of Results (One Day) This embodiment demonstrates the significant advantages of a lithium-sodium hybrid energy storage system employing dynamic reconfigurability and multi-objective adaptive optimization control in heterogeneous battery integration, dynamic energy flow allocation, and full lifecycle health management. Multivariate tensor mapping enables unified scheduling of lithium and sodium cell characteristics; topology reconfiguration allows the system to adapt to various operating conditions in real time; evolutionary optimization drives the proactive evolution of energy storage cell parameters, significantly improving overall performance and economy. These technologies can be widely applied to distributed microgrids, user-side energy storage, and new power system applications.

[0039] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Typical embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0040] Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. To facilitate understanding, the invention will now be described more fully with reference to the accompanying drawings. Typical embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to make the disclosure of the invention more thorough and complete.

Claims

1. A lithium-sodium hybrid energy storage system and grid-type control method based on dynamic reconfigurable technology, characterized in that: The method includes: The adaptive dynamic mapping of the lithium-sodium dual energy storage unit enables multi-dimensional dynamic sensing of key parameters such as real-time capacity, health status, and operating temperature of the lithium battery pack and sodium battery pack, and maps the characteristic parameters of the lithium-sodium battery to the same energy dispatch reference space. The heterogeneous energy storage unit is deeply coupled and dynamically reconfigured. According to the scheduling requirements, the interface circuit structure of the lithium-sodium battery is dynamically adjusted. The coupling strength factor is introduced. Based on energy flow, response speed and aging state, the connection mode (series, parallel, hybrid) and coupling degree of the lithium-sodium unit are determined in real time. Unified virtual synchronization control for energy storage enables multiple energy storage units to coordinate through virtual synchronization machine logic in both grid-connected and islanded operation scenarios, achieving frequency, voltage support, and inertia regulation. Based on behavioral intelligence, dynamic energy allocation and fault self-healing are achieved by learning from the historical behavioral data and abnormal patterns of energy storage units. This enables forward scheduling of energy allocation and local fault self-healing of the energy storage system. By introducing a behavioral tag library self-healing priority tree, the system can quickly identify and adjust resource scheduling strategies for atypical faults, thus maintaining stable system operation. The system performance monitoring and evolutionary adaptive optimization performs multi-dimensional real-time monitoring of the operating indicators of all system units, automatically generates the optimal combination of operating parameters based on historical evolution trends, introduces an evolutionary selection mechanism, provides parameter fine-tuning suggestions for abnormal operating conditions, and actively intervenes in scheduling strategies.

2. The lithium-sodium hybrid energy storage system and grid-type control method based on dynamic reconfigurable technology according to claim 1, characterized in that: The adaptive dynamic mapping of the lithium-sodium dual energy storage unit includes: real-time multi-dimensional acquisition of core state parameters such as capacity, health status, operating temperature, internal resistance, output voltage, and charge / discharge rate of the lithium battery pack and sodium battery pack, and normalization processing of all parameters. An adaptive weight tensor is dynamically generated based on the current reliability and priority of each energy storage unit. Through a multivariate tensor space mapping method, the normalized state parameters of the two heterogeneous energy storage units, lithium battery and sodium battery, are fused with the adaptive weights to form a unified mapping reference space vector, thereby realizing full-state adaptive mapping of different types of batteries in the energy storage system.

3. The lithium-sodium hybrid energy storage system and grid-type control method based on dynamic reconfigurable technology according to claim 1, characterized in that: The aforementioned deep coupling dynamic reconfiguration of heterogeneous energy storage units includes: using the unified mapping vector output by the Adaptive Energy Storage Mapping Algorithm (ASMA) as the global state basis, and dynamically calculating the comprehensive coupling strength factor based on energy flow capacity, response speed, and aging state parameters; Based on the real-time changes of this factor, the reconfiguration switch matrix is ​​controlled to automatically switch the series, parallel, or hybrid interface topology of lithium battery and sodium battery energy storage units. This enables deep coupling and dynamic reconfiguration of lithium-sodium heterogeneous energy storage units, effectively improving the energy efficiency, safety, and system stability of hybrid energy storage systems under various operating conditions.

4. The lithium-sodium hybrid energy storage system and grid-type control method based on dynamic reconfigurable technology according to claim 1, characterized in that: The unified virtual synchronous control of energy storage includes: real-time acquisition of multiple attribute state parameters of each energy storage unit in the lithium-sodium hybrid energy storage system, such as energy margin, output power, health status, and remaining lifespan, to form a decision input vector, and calculation of the comprehensive response capability index of each energy storage unit through a multi-attribute weighted decision function; The maximum response capability optimization method is adopted to automatically select the dominant response energy storage unit to participate in frequency and inertia support. Based on this, the main control parameters of the virtual synchronous machine, including virtual inertia and damping, are dynamically adjusted to achieve adaptive dynamic adjustment of frequency, voltage and inertia in multiple scenarios. When the system state changes or is abnormal, the response unit and weight are automatically switched to ensure the continuous and reliable operation of the system.

5. The lithium-sodium hybrid energy storage system and grid-type control method based on dynamic reconfigurable technology according to claim 1, characterized in that: The aforementioned dynamic energy allocation and fault self-healing based on behavioral intelligence includes: establishing a behavioral tag library for each energy storage unit, combining historical operating condition tags, load response tags, charging and discharging behavior tags and abnormal alarm tags, extracting behavioral feature vectors through a self-learning mechanism, and dynamically identifying operating modes and abnormal features; By utilizing a behavioral intelligence weighted inference model, behavioral-driven weighted allocation factors are calculated in real time based on behavioral labels and behavioral feature vectors to guide energy allocation optimization and achieve priority allocation of units with high reliability, low anomaly rate, and rapid response capability. Meanwhile, based on the self-healing priority tree, combined with the unit health status and historical anomalies, the system intelligently selects and switches the optimal self-healing measures, dynamically adjusts the allocation factor and output power, and realizes resource redundancy scheduling and fault self-healing of the energy storage system.

6. The lithium-sodium hybrid energy storage system and grid-type control method based on dynamic reconfigurable technology according to claim 1, characterized in that: The system performance monitoring and evolution adaptive optimization includes: real-time monitoring and analysis of efficiency, response speed, aging rate, and economic multi-dimensional performance indicators of each unit of the energy storage system; establishing a set of operating indicators and historical performance trajectories; and using a standardized comprehensive evaluation function to quantify the multi-dimensional performance. Evolutionary optimization of operating parameters is achieved through parameter population and genetic operators to maximize the overall system performance; based on performance trajectory difference and adaptive sensitivity threshold, abnormal operating conditions are automatically detected and parameter fine-tuning suggestions are generated and pushed to the scheduling center to achieve proactive intervention and optimization adjustment.

7. The lithium-sodium hybrid energy storage system and grid-type control method based on dynamic reconfigurable technology according to claim 1, characterized in that: The coupling strength factor used is calculated based on the energy flow rate of the energy storage unit, the current load demand, and the aging parameters to optimize the energy distribution ratio between lithium and sodium.

Images