High-voltage integrated and cooperative control method and system for energy storage system

Abstract

The invention discloses a high-voltage integration and cooperative control method and system for an energy storage system, and the method comprises the steps: carrying out the three-dimensional space coupling design of a high-voltage box and a power conversion system PCS through a topological optimization algorithm according to the power level and heat distribution demands of the energy storage system, and generating a physical layer architecture scheme of a high-voltage all-in-one machine; based on a physical layer architecture scheme of the high-voltage all-in-one machine, constructing a collaborative optimization data set by using a timestamp alignment technology; inputting the collaborative optimization data set into a model predictive control framework to generate dynamically adjusted PCS working parameters; and according to the PCS working parameters, the internal temperature distribution of the high-voltage all-in-one machine is calculated in real time through the three-dimensional thermal field reconstruction model, the rotating speed of a cooling fan and the switching frequency of a power device are dynamically adjusted, and a cooperative control instruction is generated. According to the embodiment of the invention, the integration level and the cooperative control capability of the high-voltage energy storage system can be improved, and the efficient and stable operation of the energy storage system is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy storage, and particularly relates to a high-voltage integration and cooperative control method and system for an energy storage system. Background Art

[0002] With the rapid development of new energy power generation and the increasing demand of the power system for flexible regulation capabilities, the energy storage system, as an important means to achieve electric energy storage and allocation, has become a key component in modern power systems. Especially for high-efficiency integration and cooperative control on the high-voltage side, it can not only significantly improve the operation efficiency of energy storage devices, but also enhance the stability and response ability of the power system. Traditional energy storage systems usually adopt a decentralized or isolated design method, managing high-voltage conversion equipment, cooling systems, and energy storage units as independent subsystems. In practical applications, due to the complex system structure and the coupling of heat dissipation and electrical connections, it is difficult to achieve an efficient spatial layout and thermal management, resulting in insufficient energy conversion efficiency, increased equipment heat loss, and greater maintenance difficulty. Summary of the Invention

[0003] The purpose of the present invention is to provide a high-voltage integration and cooperative control method and system for an energy storage system to solve the deficiencies in the prior art, improve the integration level and cooperative control ability of the high-voltage energy storage system, and achieve the efficient and stable operation of the energy storage system.

[0004] An embodiment of the present application provides a high-voltage integration and cooperative control method for an energy storage system, and the method includes: According to the power level and heat distribution requirements of the energy storage system, a three-dimensional space coupling design of the high-voltage box and the power conversion system PCS is carried out through a topology optimization algorithm to generate a physical layer architecture scheme of the high-voltage integrated machine, wherein the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel; Based on the physical layer architecture scheme of the high-voltage integrated machine, the real-time battery state data of the battery management system BMS and the grid dispatching instructions of the energy management system EMS are synchronously collected through a dynamic data bus, and a cooperative optimization data set is constructed by using timestamp alignment technology, wherein the cooperative optimization data set includes multi-dimensional coupling characteristics such as battery health, charge and discharge rate, and grid demand response; The cooperative optimization data set is input into the model predictive control framework, and a rolling horizon optimization algorithm is used to solve the charge and discharge strategy to generate dynamically adjusted PCS working parameters, wherein the rolling horizon optimization algorithm realizes multi-objective cooperative optimization by restricting the battery life attenuation rate and the grid power fluctuation range; According to the PCS working parameters, the internal temperature distribution of the high-voltage integrated machine is calculated in real time through a three-dimensional thermal field reconstruction model, the rotation speed of the cooling fan and the switching frequency of the power device are dynamically adjusted to generate a collaborative control instruction, and at the same time, it is fed back to the BMS and EMS to update the battery charge and discharge thresholds and the grid dispatching strategy, forming a closed-loop control link.

[0005] Optionally, according to the power level and thermal distribution requirements of the energy storage system, the high-voltage box and the power conversion system PCS are subjected to three-dimensional space coupling design through a topology optimization algorithm to generate a physical layer architecture scheme of the high-voltage integrated machine. Among them, the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel, including: According to the power level parameters and thermal distribution requirements of the energy storage system, a three-dimensional initial model of the high-voltage box and PCS is constructed to generate an initial three-dimensional grid model including electrical interfaces, heat dissipation port positions and internal cavities; In the initial three-dimensional grid model, the electrical connection path is designed based on the minimum path principle, and the length and impedance of the busbar wiring are iteratively optimized through the ant colony algorithm to generate a candidate set of electrical connection paths and mark the key node coordinates; At the same time, a hydrodynamic simulation is carried out on the heat dissipation channel, the layout of the heat dissipation fins is optimized according to the heat source distribution density, the heat dissipation efficiency curve at different wind speeds is calculated, and the topological structure of the heat dissipation channel and the fan deployment scheme are generated; The candidate set of electrical connection paths and the topological structure of the heat dissipation channel are input into a multi-objective optimization algorithm. With the electrical loss rate and heat dissipation efficiency as the constraint conditions, the Pareto optimal solution set is solved to generate a topological scheme of electrical-heat dissipation coupling; An electromagnetic-thermal coupling simulation verification is carried out on the topological scheme, local hot spots and areas with excessive electromagnetic interference are detected, the connection path and the heat dissipation channel spacing are corrected, and the final three-dimensional physical layer architecture scheme of the high-voltage integrated machine is output.

[0006] Optionally, based on the physical layer architecture of the high-voltage integrated machine, the real-time battery state data of the battery management system BMS and the grid dispatching instructions of the energy management system EMS are synchronously collected through a dynamic data bus, and a collaborative optimization data set is constructed by using the timestamp alignment technology. Among them, the collaborative optimization data set includes multi-dimensional coupling characteristics such as battery health, charge and discharge rate, and grid demand response, including: Collect the battery state data of the BMS and the grid dispatching instructions of the EMS through the dynamic data bus, and use the hardware clock synchronization technology to align the data timestamps to generate a time-synchronized original data stream; Clean the outliers of the original data stream, compensate for the missing data based on the battery cluster topology relationship, and eliminate the sensor noise interference by using the covariance matrix to generate a purified multi-dimensional monitoring data set; Extract battery health characteristics, charge and discharge rate characteristics, and grid demand response characteristics from the multi-dimensional monitoring dataset to construct a multi-dimensional coupling feature correlation matrix; According to the multi-dimensional coupling feature correlation matrix, analyze the dynamic correlation between battery health, charge and discharge rate, and grid demand response pairwise through the Pearson correlation coefficient, and generate a multi-dimensional coupling feature mapping table containing weight labels; Fuse the multi-dimensional coupling feature mapping table with time series data, and package it into a collaborative optimization dataset with spatio-temporal labels according to a preset sampling period, and store it in the shared memory area of the edge computing node.

[0007] Optionally, input the collaborative optimization dataset into the model predictive control framework, and use the rolling horizon optimization algorithm to solve the charge and discharge strategy to generate dynamically adjusted PCS operating parameters. Among them, the rolling horizon optimization algorithm realizes multi-objective collaborative optimization by constraining the battery life attenuation rate and the grid power fluctuation range, including: Input the collaborative optimization dataset into the model predictive control framework, predict the power demand curve for a preset number of future control cycles based on the current battery state and grid instructions, and generate a reference trajectory for the charge and discharge power benchmark; Establish a multi-objective optimization model, refer to the reference trajectory of the charge and discharge power benchmark, use the battery life attenuation rate as the first constraint and the grid power fluctuation range as the second constraint, and define the objective function as the maximization of system efficiency; Use the rolling horizon optimization algorithm to solve the objective function, iteratively calculate the optimal charge and discharge current values, PCS switching frequency, and power factor compensation parameters in each control cycle, and generate a candidate set of dynamic PCS operating parameters; Verify the feasibility of the candidate parameters in the candidate set of dynamic PCS operating parameters, detect whether the parameters exceed the safety boundary through the battery polarization voltage model and the grid harmonic analysis model, and output a set of dynamically adjusted PCS operating parameter instructions that pass the verification.

[0008] Optionally, according to the PCS operating parameters, calculate the internal temperature distribution of the high-voltage integrated machine in real time through a three-dimensional thermal field reconstruction model, dynamically adjust the rotation speed of the cooling fan and the switching frequency of the power device, generate a collaborative control instruction, and at the same time feedback to the BMS and EMS to update the battery charge and discharge threshold and the grid dispatching strategy to form a closed-loop control link, including: According to the PCS operating parameters, calculate the power device loss value in real time by the finite element method, and construct a three-dimensional thermal field distribution model inside the high-voltage integrated machine in combination with the fluid mechanics equation; Identify the area with excessive temperature in the three-dimensional thermal field distribution model, dynamically adjust the rotation speed gear of the cooling fan and the switching frequency of the power device, and generate a draft of the heat dissipation control instruction; Input the draft of the heat dissipation control instruction into the digital twin system for simulation execution, predict the temperature change trend after adjustment. If the simulation result does not meet the expectation, trigger instruction replanning until the thermal management requirements are met, and obtain the final collaborative control instruction. Send the final collaborative control instruction to the PCS and the heat dissipation system for execution. At the same time, feedback the temperature distribution data to the BMS to update the battery charge and discharge threshold, and send a correction suggestion for the grid dispatching strategy to the EMS to form a closed-loop control link.

[0009] Another embodiment of the present application provides an energy storage system high-voltage integration and collaborative control system, and the system includes: A design module, configured to perform three-dimensional space coupling design on the high-voltage box and the power conversion system PCS according to the power level and heat distribution requirements of the energy storage system through a topology optimization algorithm, and generate a physical layer architecture scheme of the high-voltage integrated machine. Among them, the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel. A construction module, configured to synchronously collect the real-time battery state data of the battery management system BMS and the grid dispatching instruction of the energy management system EMS based on the physical layer architecture scheme of the high-voltage integrated machine through a dynamic data bus, and construct a collaborative optimization data set by using the timestamp alignment technology. Among them, the collaborative optimization data set includes multi-dimensional coupling characteristics such as battery health, charge and discharge rate, and grid demand response. A generation module, configured to input the collaborative optimization data set into a model predictive control framework, solve the charge and discharge strategy by using a rolling horizon optimization algorithm, and generate dynamically adjusted PCS operating parameters. Among them, the rolling horizon optimization algorithm realizes multi-objective collaborative optimization by restricting the battery life attenuation rate and the grid power fluctuation range. A control module, configured to calculate the internal temperature distribution of the high-voltage integrated machine in real time through a three-dimensional thermal field reconstruction model according to the PCS operating parameters, dynamically adjust the rotation speed of the heat dissipation fan and the switching frequency of the power device, generate a collaborative control instruction, and at the same time feedback to the BMS and the EMS to update the battery charge and discharge threshold and the grid dispatching strategy to form a closed-loop control link.

[0010] Another embodiment of the present application provides a storage medium, in which a computer program is stored. Among them, the computer program is set to execute the method described in any one of the above when running.

[0011] Another embodiment of the present application provides an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is set to run the computer program to execute the method described in any one of the above.

[0012] Compared with the prior art, a high-voltage integration and collaborative control method for an energy storage system provided by the present invention designs the three-dimensional space coupling of a high-voltage box and a power conversion system (PCS) through a topology optimization algorithm according to the power level and thermal distribution requirements of the energy storage system, and generates a physical layer architecture scheme of a high-voltage integrated machine; based on the physical layer architecture scheme of the high-voltage integrated machine, a collaborative optimization data set is constructed by using a timestamp alignment technology; the collaborative optimization data set is input into a model predictive control framework to generate dynamically adjusted PCS operating parameters; according to the PCS operating parameters, the internal temperature distribution of the high-voltage integrated machine is calculated in real time through a three-dimensional thermal field reconstruction model, and the rotation speed of the cooling fan and the switching frequency of the power device are dynamically adjusted to generate collaborative control instructions, so as to improve the integration level and collaborative control ability of the high-voltage energy storage system and realize the efficient and stable operation of the energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 FIG. is a hardware structure block diagram of a computer terminal for a high-voltage integration and collaborative control method of an energy storage system provided by an embodiment of the present invention; Figure 2 FIG. is a schematic flowchart of a high-voltage integration and collaborative control method of an energy storage system provided by an embodiment of the present invention; Figure 3 FIG. is a schematic structural diagram of a high-voltage integration and collaborative control system of an energy storage system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and should not be construed as limiting the present invention.

[0015] An embodiment of the present invention first provides a high-voltage integration and collaborative control method for an energy storage system, which can be applied to an electronic device, such as a computer terminal, specifically, a general computer, etc.

[0016] The following will describe it in detail by taking the operation on a computer terminal as an example. Figure 1 FIG. is a hardware structure block diagram of a computer terminal for a high-voltage integration and collaborative control method of an energy storage system provided by an embodiment of the present invention. As Figure 1 shown, the computer device includes a processor, a memory, and a network interface connected through a system bus. Among them, the memory may include a non-volatile storage medium and an internal memory.

[0017] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any high-voltage integration and collaborative control method of an energy storage system.

[0018] The processor is used to provide computing and control capabilities to support the operation of the entire computer device.

[0019] The internal memory provides an environment for the operation of a computer program in a non-volatile storage medium. When the computer program is executed by a processor, the processor can be caused to execute any one of the high-voltage integration and coordinated control methods for an energy storage system.

[0020] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art can understand that Figure 1 The structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0021] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0022] See Figure 2 , the embodiments of the present invention provide a high-voltage integration and coordinated control method for an energy storage system, which may include the following steps: S201. According to the power level and thermal distribution requirements of the energy storage system, perform three-dimensional space coupling design on the high-voltage box and the power conversion system PCS through a topology optimization algorithm to generate a physical layer architecture scheme for the high-voltage integrated machine, where the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel; specifically, it may include: According to the power level parameters and thermal distribution requirements of the energy storage system, construct a three-dimensional initial model of the high-voltage box and the PCS, and generate an initial three-dimensional grid model including electrical interfaces, heat dissipation port positions, and internal cavities; Analysis of Power Level and Thermal Distribution Requirements The power level parameters of the energy storage system include rated voltage (such as 1500V), maximum charge and discharge current (such as 300A), and peak power (such as 1MW). The thermal distribution requirements are determined according to the battery cluster layout and the thermal dissipation characteristics of PCS power devices (such as IGBT modules). For example, it is required that the temperature rise inside the high-voltage box does not exceed 40°C (the maximum temperature is 65°C when the ambient temperature is 25°C).

[0023] Heat Source Location: Mark the main heat-generating areas through thermal imaging data, such as the DC / AC module of the PCS (coordinates x = 1.2m, y = 0.8m) and the bus connection point of the high-voltage box (coordinates x = 0.5m, y = 1.0m).

[0024] 3D Modeling Tools: Use SolidWorks or ANSYS SpaceClaim to build the geometric models of the high-voltage box and the PCS, and define the electrical interfaces (such as DC input terminals, AC output terminals) and the positions of the heat dissipation ports (size 200mm×100mm, located at the top of the box).

[0025] Generation of the Initial 3D Mesh Model Mesh Generation: Discretize the model using tetrahedral elements. The mesh size is set from 5mm (for critical areas such as heat dissipation ports) to 20mm (for non-critical areas) according to the accuracy requirements. For example, the internal cavity of the high-voltage box is divided into 500,000 mesh nodes, and the PCS module is divided into 300,000 nodes.

[0026] Mesh Optimization: Detect deformed meshes (such as elements with an aspect ratio > 10) through the Mesh Quality Check function, and use the Laplacian smoothing algorithm to optimize the mesh quality to ensure calculation convergence.

[0027] In the initial 3D mesh model, design the electrical connection path based on the minimum path principle. Iteratively optimize the length and impedance of the busbar routing through the ant colony algorithm, generate a candidate set of electrical connection paths, and mark the coordinates of the key nodes; Minimum Path Principle and Parameter Settings of the Ant Colony Algorithm Path Constraint Conditions: The busbar routing needs to meet the safety distance (such as the distance from the heat dissipation channel ≥ 10mm), bending radius (≥ 5 times the busbar thickness), and impedance limit (≤ 0.1mΩ / m).

[0028] Ant Colony Algorithm Parameters: Set the pheromone evaporation rate to 0.3, the heuristic factor α = 1, β = 2, and the number of iterations to 500 times.

[0029] Mother bus path optimization process Initial path generation: Randomly generate 10 initial paths from the DC input end of the high-voltage box (starting coordinates x = 0, y = 0) to the PCS module (ending coordinates x = 1.2m, y = 0.8m), with a length range of 1.5m - 2.0m and an impedance range of 0.15mΩ - 0.25mΩ.

[0030] Iterative optimization: Each "ant" selects the next node based on the pheromone concentration and path length. For example, after the 100th iteration, the optimal path length is shortened to 1.2m and the impedance is reduced to 0.08mΩ.

[0031] Candidate set generation: Retain the top 5 optimal paths (such as lengths 1.2m, 1.3m, 1.4m, and impedances 0.08mΩ, 0.09mΩ, 0.10mΩ), and mark the key nodes (such as the inflection point coordinates x = 0.8m, y = 0.6m).

[0032] Synchronously perform a hydrodynamic simulation on the heat dissipation channel, optimize the layout of the heat dissipation fins according to the heat source distribution density, calculate the heat dissipation efficiency curve at different wind speeds, and generate the topological structure of the heat dissipation channel and the fan deployment plan; Hydrodynamic simulation and heat dissipation fin optimization Calculation of heat source distribution density: Determine the layout density of the heat dissipation fins based on the heat dissipation data of the three-dimensional grid model (such as the heat flux density of the PCS module is 5000W / m², and the heat flux density of the bus connection point of the high-voltage box is 3000W / m²).

[0033] Heat dissipation fin parameters: The fin height is 30mm, the thickness is 2mm, the spacing is 10mm, and the material is aluminum alloy (thermal conductivity 237W / m·K). Through ANSYS Fluent simulation, compare the heat dissipation efficiency of different layouts (such as parallel arrangement and staggered arrangement).

[0034] Wind speed - heat dissipation efficiency curve: The heat dissipation efficiency is 60% (temperature rise reduction of 24℃) at a wind speed of 2m / s, increases to 75% (temperature rise reduction of 30℃) at a wind speed of 4m / s, and the efficiency growth slows down after the wind speed exceeds 6m / s (diminishing marginal benefit).

[0035] Design of fan deployment plan Fan selection: Select an axial fan (such as the EBM-papst model A3G450, air volume 200CFM, rotational speed 2500RPM), and determine the deployment quantity according to the heat dissipation efficiency curve (such as 4 fans, located at the top and both sides of the box respectively).

[0036] Topological structure output: The flow channel width of the heat dissipation channel is set to 50 mm, and the fan installation coordinates (such as the top fan coordinates x = 0.5 m, y = 1.5 m, and the side fan coordinates x = 0.2 m, y = 0.5 m and x = 1.8 m, y = 0.5 m).

[0037] Input the candidate set of electrical connection paths and the heat dissipation channel topological structure into the multi-objective optimization algorithm. With the electrical loss rate and heat dissipation efficiency as the constraint conditions, solve the Pareto optimal solution set to generate an electrically-heat dissipation coupled topological scheme; Multi-objective optimization model construction Objective function: Minimize the electrical loss rate (objective value ≤ 0.1 mΩ) and maximize the heat dissipation efficiency (objective value ≥ 70%).

[0038] Constraint conditions: The length of the busbar path ≤ 1.5 m, the flow channel wind speed of the heat dissipation channel ≥ 3 m / s, and the fan power consumption ≤ 200 W.

[0039] Solve the Pareto optimal solution by NSGA-II algorithm Population initialization: Generate 100 individuals, each individual contains a busbar path and a heat dissipation topology combination. For example, individual 1 has a path length of 1.2 m and a heat dissipation efficiency of 75%, and individual 2 has a path length of 1.3 m and a heat dissipation efficiency of 80%.

[0040] Iterative evolution: After 200 generations of evolution, the Pareto front contains 20 non-dominated solutions, and the optimal solution is a path length of 1.25 m, an impedance of 0.09 mΩ, and a heat dissipation efficiency of 78%.

[0041] Scheme selection: According to engineering requirements (such as cost constraints), select the solution with priority on heat dissipation efficiency: a path length of 1.3 m, an impedance of 0.10 mΩ, a heat dissipation efficiency of 80%, and the number of fans is reduced to 3.

[0042] Conduct electromagnetic-thermal coupling simulation verification on the topological scheme, detect local hot spots and areas with excessive electromagnetic interference, correct the connection path and the spacing of the heat dissipation channel, and output the three-dimensional scheme of the physical layer architecture of the final high-voltage integrated machine.

[0043] Electromagnetic-thermal coupling simulation execution Simulation tool: Use COMSOL Multiphysics, load the three-dimensional model of the topological scheme, and set the boundary conditions (such as an environmental temperature of 25 °C and a busbar current of 300 A).

[0044] Hot Spot Detection: The simulation results show that the temperature in the area below the PCS module (coordinates x = 1.2m, y = 0.8m, z = 0.5m) reaches 68°C (exceeding the threshold of 65°C), and the magnetic field intensity at the inflection point of the busbar (coordinates x = 0.8m, y = 0.6m) exceeds the standard (>50μT).

[0045] Scheme Revision and Verification Spacing Adjustment: Increase the spacing between the busbar and the heat dissipation channel from 10mm to 15mm, and add an electromagnetic shielding layer (material: MuMetal, thickness: 1mm) at the inflection point of the busbar.

[0046] Secondary Simulation: After adjustment, the maximum temperature drops to 63°C, and the magnetic field intensity drops to 45μT, meeting the design requirements.

[0047] Final Output: Generate a 3D model file in STEP format, including the modified geometric structure, material properties, and performance parameters (such as electrical loss rate 0.10mΩ, heat dissipation efficiency 80%).

[0048] Technical Details and Parameter Examples 3D Mesh Model Parameters Total number of meshes: 800,000 (high-voltage box) + 500,000 (PCS) = 1.3 million elements; Minimum mesh size: 5mm (for busbar connection points and heat dissipation fins); Computing Resources: The simulation requires 64GB of memory and takes 6 hours (based on an Intel Xeon 16-core processor).

[0049] Optimization Effect of Ant Colony Algorithm Average length of the initial path: 1.8m, shortest path after optimization: 1.2m; Reduction amplitude of impedance: Optimized from 0.25mΩ to 0.08mΩ (a reduction of 68%).

[0050] Relationship between Heat Dissipation Efficiency and Wind Speed Heat dissipation power at a wind speed of 2m / s: 5000W; Heat dissipation power at a wind speed of 4m / s: 8000W; Critical wind speed: 6m / s (continuing to increase the wind speed only increases the heat dissipation power to 8200W).

[0051] Comparison of Multi-objective Optimization Results Pareto solution 1: Loss rate 0.09mΩ, heat dissipation efficiency 78%, cost $1200; Pareto solution 2: Loss rate 0.10mΩ, heat dissipation efficiency 80%, cost $1000; Final selected solution: Solution 2 (meeting the cost constraint).

[0052] Application Scenario Example Scenario Description: Design of a high-voltage integrated machine for a 1500V / 1MW energy storage system

[0053] 3D Modeling: The size of the high-voltage box is 2m×1m×1.5m, and the size of the PCS module is 0.8m×0.6m×0.4m; Electrical Path Optimization: The length of the busbar is optimized from 1.8m to 1.3m, and the impedance is 0.10mΩ; Thermal Design: Deploy 3 axial fans, and the heat dissipation efficiency is 80%; Coupled Simulation Verification: The maximum temperature is 63°C, and the magnetic field intensity is 45μT; Final Solution: Submit the 3D model file and performance report, and pass the customer acceptance test.

[0054] This step is based on the power level (such as hundreds of kilowatts or megawatts) of the energy storage system and the thermal distribution requirements (such as the heat generation density of power devices), and conducts a 3D space coupling design of the high-voltage box and the power conversion system (PCS) through a topology optimization algorithm. Specifically, it includes: using the ant colony algorithm to optimize the busbar routing path to reduce electrical impedance, and at the same time combining computational fluid dynamics (CFD) simulation to optimize the layout of heat dissipation fins to improve heat dissipation efficiency, and balancing electrical losses and heat dissipation performance through a multi-objective optimization algorithm (such as Pareto optimality), and finally generating a 3D physical architecture solution for the high-voltage integrated machine to ensure that there are no conflicts in space between the electrical connection path and the heat dissipation channel and the efficiency is maximized, solving the problems of volume redundancy and low efficiency caused by the separation of electrical and thermal designs in traditional energy storage systems, and providing a compact and integrated hardware foundation for high-power density energy storage systems.

[0055] S202, based on the physical layer architecture solution of the high-voltage integrated machine, synchronously collect the real-time battery state data of the battery management system BMS and the grid dispatching instructions of the energy management system EMS through the dynamic data bus, and use the timestamp alignment technology to construct a collaborative optimization data set, where the collaborative optimization data set includes multi-dimensional coupling characteristics such as battery health, charge and discharge rate, and grid demand response; specifically, it can include: Collect the battery state data of the BMS and the grid dispatching instructions of the EMS through the dynamic data bus, and use the hardware clock synchronization technology to align the data timestamps to generate a time-synchronized original data stream; Selection and Configuration of the Dynamic Data Bus The dynamic data bus adopts the EtherCAT (Ethernet Control Automation Technology) or CAN FD (Controller Area Network Flexible Data Rate) protocol, supporting high-speed (≥100Mbps) and low-latency (<1ms) data transmission. For example, the BMS sends data such as battery cell voltage (accuracy ±1mV), temperature (accuracy ±0.5°C), and SOC (State of Charge, accuracy ±2%) through the CAN FD bus at a frequency of 1kHz, and the EMS sends grid dispatching instructions (such as charge and discharge power instructions, frequency modulation signals, peak-valley electricity price information) through the EtherCAT bus.

[0056] ‌Bus topology design‌: Adopt a star topology structure, with edge computing nodes (such as NVIDIA Jetson AGXXavier) as the master station, and the BMS and EMS as slave stations. The polling period of the master station is set to 2ms to ensure data real-time performance.

[0057] ‌Hardware clock synchronization‌: Based on the IEEE 1588 PTP (Precision Time Protocol), the master station and slave stations deploy hardware clock synchronization modules (such as Texas Instruments DP83640 chips) to calibrate the timestamp accuracy to 0.1ms. For example, the time deviation between the BMS and EMS is reduced from the initial 5ms to 0.2ms.

[0058] ‌Timestamp alignment and data stream generation‌ ‌Timestamp marking‌: Each data packet embeds the local clock timestamp at the sending end (such as Unix timestamp format: 1630453200.123456), and synchronizes the master station clock to the slave stations through the PTP protocol.

[0059] ‌Alignment algorithm‌: Adopt the ‌interpolation method‌ to compensate for clock deviation. For example, the voltage data sent by the BMS at timestamp t = 1000ms and the dispatching instruction sent by the EMS at t = 1000.2ms generate an aligned data point t = 1000.1ms through linear interpolation.

[0060] ‌Original data stream format‌: Generate a structured data stream containing aligned timestamps (Timestamp), BMS data (Voltage, Temperature, SOC), and EMS instructions (Power_Setpoint, Frequency_Command), and store it in CSV or binary format.

[0061] ‌Perform outlier cleaning on the original data stream, compensate for missing data based on the battery cluster topology relationship, and use the covariance matrix to eliminate sensor noise interference to generate a purified multi-dimensional monitoring data set; ‌Outlier detection and cleaning‌ Rule base filtering: Define threshold rules (e.g., abnormal when the single cell voltage < 2.5V or > 4.2V, abnormal when the temperature < -20°C or > 60°C), and automatically eliminate data points outside the range.

[0062] Statistical method: Use the Z-score algorithm to detect outliers. For example, calculate the mean and standard deviation of voltage data (e.g., μ = 3.6V, σ = 0.1V), and a Z-score > 3 (i.e., voltage > 3.9V or < 3.3V) is determined to be abnormal.

[0063] Sliding window repair: For continuous abnormal data (e.g., the voltage of a certain battery single cell is abnormal for 3 consecutive cycles), replace it with the median of the adjacent window (window size = 10 sampling points).

[0064] Missing data compensation Battery cluster topology relationship modeling: Based on the series-parallel structure of the battery cluster (e.g., 16S48P: 16 series modules, each module has 48 parallel single cells), establish an electrical association model between single cells.

[0065] Compensation strategy: If the data of a certain single cell is missing (e.g., the voltage of ID = 5 is lost), fill it with the average value of other single cells in the same module (e.g., the average voltage of 47 single cells in the module is 3.65V).

[0066] Time series interpolation: For data with short-term missing (e.g., < 5 sampling points), use cubic spline interpolation to complete it.

[0067] Sensor noise suppression Covariance matrix construction: Assume that the noise follows a Gaussian distribution, and construct the covariance matrix (Covariance Matrix) of voltage, temperature, and SOC. For example: Cov(Voltage, Temperature) = -0.05 (negative correlation: the voltage decreases as the temperature increases); Cov(SOC, Voltage) = 0.8 (strong positive correlation: the voltage is high when the SOC is high).

[0068] Kalman filter noise reduction: Design a Kalman filter based on the covariance matrix to predict the true state and correct the noise. For example, the measured voltage value is 3.7V ± 0.1V, and the output after filtering is 3.68V ± 0.03V.

[0069] Extract battery health characteristics, charge and discharge rate characteristics, and grid demand response characteristics from the multi-dimensional monitoring dataset to construct a multi-dimensional coupling feature correlation matrix; Battery health (SOH) feature extraction Calculation of capacity attenuation rate: Based on historical charge and discharge data (for example, the capacity drops from 100 Ah to 85 Ah after 1000 cycles), calculate SOH = current capacity / initial capacity × 100% (SOH = 85%).

[0070] Growth characteristics of internal resistance: Obtain the internal resistance value through HPPC (Hybrid Pulse Power Characterization) test (for example, the initial internal resistance R0 = 1 mΩ, the current internal resistance R = 1.5 mΩ), and the internal resistance growth rate ΔR = (R - R0) / R0 × 100% = 50%.

[0071] Capacity-internal resistance coupling index: Define the comprehensive SOH score = 0.7 × capacity attenuation rate + 0.3 × internal resistance growth rate (weights are calibrated according to experiments). For example, SOH score = 0.7 × 85% + 0.3 × 50% = 74.5%.

[0072] Feature extraction of charge and discharge rate (C-rate) Calculation of real-time C-rate: C-rate = charge and discharge current (A) / rated capacity (Ah). For example, when the current is 300 A and the rated capacity is 100 Ah, then C-rate = 3C.

[0073] C-rate change rate: Calculate the C-rate change gradient (ΔC-rate / Δt) in adjacent time windows (such as 1 second). For example, when it rises from 2C to 3C, the gradient = 1C / s.

[0074] Voltage slope feature: During the constant current charge and discharge stage, the voltage change rate dV / dt reflects the polarization state of the battery. For example, during charging, dV / dt = 0.05 V / s, and during discharging, dV / dt = -0.08 V / s.

[0075] Feature extraction of grid demand response Power deviation: Calculate the difference between the EMS command power and the actual power (ΔP = Setpoint - Actual). For example, ΔP = 50 kW means that the discharge power needs to be increased by 50 kW.

[0076] Frequency modulation response delay: Statistically calculate the delay time from the command issuance to the completion of power adjustment (such as the average delay = 200 ms).

[0077] Economic index: According to the time-of-use electricity price (such as 1.2 yuan / kWh during peak hours and 0.5 yuan / kWh during valley hours), calculate the difference between the charge and discharge cost and revenue (such as the discharge revenue = 1.2 × 500 kWh = 600 yuan, the charging cost = 0.5 × 500 kWh = 250 yuan, and the net revenue = 350 yuan).

[0078] Construction of multi-dimensional coupling feature correlation matrix Matrix Dimension Design: The rows represent time series samples (e.g., one sample per second), and the columns include 10 features such as capacity attenuation rate, internal resistance growth, SOH score, C-rate, C-rate change rate, voltage slope, ΔP, frequency modulation response delay, and economic indicators.

[0079] Normalization: Min-Max normalization is performed on the feature values. For example, the SOH score range [60%, 100%] is mapped to [0, 1], and the C-rate range [0C, 5C] is mapped to [0, 1].

[0080] Matrix Storage Structure: Stored using Numpy arrays or Pandas DataFrames. For example, the matrix size of a 1-hour dataset (3600 samples) is 3600×10.

[0081] Based on the multi-dimensional coupled feature correlation matrix, analyze the dynamic correlation relationships between battery health, charge and discharge rates, and grid demand response pairwise through Pearson correlation coefficients, and generate a multi-dimensional coupled feature mapping table containing weight labels; Pearson Correlation Coefficient Calculation Calculation Steps: Calculate the covariance divided by the product of the standard deviations for each pair of features (e.g., SOH score and C-rate). For example: r(SOH, C-rate) = Cov(SOH, C-rate) / (σ_SOH × σ_C-rate). Assume Cov = -0.15, σ_SOH = 0.2, σ_C-rate = 0.3, then r = -0.15 / (0.2×0.3) = -2.5 (needs to be corrected as it exceeds the [-1, 1] range, and the actual value should be -0.25).

[0082] Dynamic Window Analysis: Calculate the real-time correlation coefficient using a sliding window (window size = 300 samples, i.e., 5 minutes of data). For example, the correlation coefficient between SOH and C-rate at t = 10:00 is -0.6, indicating that a high C-rate accelerates SOH decay.

[0083] Weight Label Assignment Correlation Coefficient Classification: Define the association strength according to the |r| value: |r|≥0.8: Strong association (weight = 1.0); 0.5≤|r|<0.8: Moderate association (weight = 0.7); |r|<0.5: Weak association (weight = 0.3).

[0084] Dynamic Weight Adjustment: For example, the correlation coefficient between SOH and the grid power deviation ΔP has a weight of 0.7 during peak hours (r = 0.7) and a weight of 0.3 during off-peak hours (r = 0.4).

[0085] Multi-dimensional Coupled Feature Mapping Table Generation Table Structure: It contains fields such as feature pair name, correlation coefficient, weight, timestamp, etc. For example, Table 1:

[0086] Real-time Update Mechanism: The mapping table is updated every 5 minutes (i.e., when the window slides once) and stored in the Redis cache database.

[0087] Fuse the multi-dimensional coupled feature mapping table with time series data, and package it into a collaborative optimization dataset with spatio-temporal tags according to the preset sampling period, and store it in the shared memory area of the edge computing node.

[0088] Spatio-temporal Tag Definition Time Tag: Adopt the ISO 8601 standard (such as 2023-10-01T10:00:00.000Z), accurate to milliseconds.

[0089] Space Tag: Encode according to the deployment location of the energy storage system, such as site ID (Site_ID=CN_BJ_001), battery cluster ID (Cluster_ID=BC_01).

[0090] Data Fusion and Packaging Sampling Period Setting: The preset sampling period is 1 second. Every second, the multi-dimensional coupled feature mapping table is merged with the original time series data (such as voltage, temperature).

[0091] Data Format Conversion: Convert to the Parquet columnar storage format, which supports efficient compression (compression ratio ≥ 5:1) and fast query.

[0092] Metadata Attachment: Add fields such as data version number (Version=1.2), checksum (CRC32), etc.

[0093] Storage in the Shared Memory Area Edge Computing Node Configuration: Adopt a multi-core ARM processor (such as NXP Layerscape LX2160A), and allocate the shared memory area size = 4GB.

[0094] Memory Database Selection: Use Apache Arrow or Redis as the memory database, which supports high-speed reading and writing (throughput ≥ 100,000 records / second).

[0095] Data Partitioning Strategy: Partition and store according to the time range (such as one partition per hour) and space tags (site ID) for parallel access.

[0096] Application Scenario Example Scenario description: Co - optimization of BMS and EMS in a 100MWh energy storage power station.

[0097] Data collection: The BMS uploads the voltage and temperature data of 192 battery cells at a frequency of 1kHz, and the EMS issues grid dispatching instructions once per second.

[0098] Data cleaning: Repair 5 abnormal voltage points (3.8V → 3.65V), and compensate for 2 missing temperature data (using the average value of 25°C in the same module).

[0099] Feature extraction: Calculate the SOH score (82%), C - rate (2.5C), and power deviation (ΔP = 30kW).

[0100] Correlation analysis: It is found that SOH and C - rate are strongly negatively correlated (r = - 0.7), triggering an optimization strategy to reduce the C - rate.

[0101] Dataset storage: Pack data in 1 - second cycles and store it in the memory of the edge node for real - time control algorithm calls.

[0102] Real - time collect the battery cell voltage, temperature, SOC data of the BMS and the grid frequency modulation and peak shaving instructions of the EMS through a dynamic data bus (such as CAN or EtherCAT), use the hardware clock synchronization technology to align the millisecond - level timestamps, and eliminate the sensor noise interference through the covariance matrix. Extract features such as battery health (such as capacity attenuation rate), charge - discharge rate (such as C - rate fluctuation), and grid demand response (such as frequency modulation power change rate), construct a multi - dimensional coupling feature matrix, form a collaborative optimization dataset containing spatio - temporal tags, break the data islands between the BMS and the EMS, achieve millisecond - level cross - system data synchronization and feature fusion, provide high - precision input for subsequent model predictive control, and avoid inaccurate control instructions caused by data delay or noise.

[0103] S203, input the collaborative optimization dataset into the model predictive control framework, use the rolling - horizon optimization algorithm to solve the charge - discharge strategy, and generate dynamically adjusted PCS operating parameters. Among them, the rolling - horizon optimization algorithm realizes multi - objective collaborative optimization by restricting the battery life attenuation rate and the grid power fluctuation range; specifically, it may include: Input the collaborative optimization dataset into the model predictive control framework, predict the power demand curve for a preset number of future control cycles based on the current battery state and grid instructions, and generate a reference trajectory for the charge - discharge power benchmark. Model Predictive Control (MPC) framework construction Input data preprocessing: The collaborative optimization dataset contains features such as State of Health (SOH) of the battery, charge and discharge rate (C-rate), and grid power deviation (ΔP). For example, SOH = 85%, C-rate = 2.5C, ΔP = 30kW. The data is normalized (Min-Max normalized to the range [0,1]) by an edge computing node (such as an Intel Xeon D-2145NT processor), and the training set (80%) and test set (20%) are divided.

[0104] Prediction model selection: The Long Short-Term Memory network (LSTM) or Autoregressive Integrated Moving Average model (ARIMA) is used to predict future power demand. For example, the LSTM network is configured with 3 hidden layers (128 neurons per layer), a time window length of 10 control cycles (each cycle is 5 seconds), and a prediction step of 3 cycles (15 seconds).

[0105] Power demand curve generation: Based on the current battery State of Charge (SOC) (such as 60%) and the EMS scheduling instruction (such as a discharge power demand of 500kW), the power change every 5 seconds within the next 15 seconds is predicted. For example, the prediction result: the discharge power needs to increase to 520kW at t = 5s and decrease to 490kW at t = 10s, forming a reference trajectory.

[0106] Dynamic reference adjustment mechanism Real-time feedback correction: At the end of each control cycle (such as 5 seconds), the actual power is compared with the predicted value, and the residual error is calculated. If the residual exceeds the threshold (such as ±5%), online update of the model parameters is triggered. For example, the actual discharge power is 515kW (predicted value 520kW), the residual = -5kW (-1%), and correction is not triggered; if the residual reaches -25kW (-5%), the LSTM weights are adjusted by the Recursive Least Squares (RLS) method.

[0107] Multi-scenario prediction fusion: For different scenarios such as grid frequency regulation and peak shaving, multiple reference trajectories are generated (for example, the frequency regulation scenario requires rapid power fluctuations, and the peak shaving scenario requires smooth changes), and the optimal trajectory is selected through weighted voting. For example, the weight for the frequency regulation scenario = 0.7, the weight for the peak shaving scenario = 0.3, and the final trajectory is a linear combination of the two.

[0108] Establish a multi-objective optimization model, referring to the reference trajectory of the charge and discharge power benchmark, with the battery life attenuation rate as the first constraint and the grid power fluctuation range as the second constraint, and define the objective function as maximizing the system efficiency; Definition of multi-objective optimization model Objective Function Design: System Efficiency = Charge-Discharge Efficiency × Grid Response Efficiency × Battery Health Factor. For example: Charge-Discharge Efficiency = Actual Output Energy / Battery Released Energy (typical value 95%); Grid Response Efficiency = 1 - |ΔP| / Setpoint (ΔP = Actual Power Deviation); Battery Health Factor = 1 - 0.1×(1 - SOH) (when SOH = 85%, the factor = 0.985).

[0109] Objective Function = 0.95×0.98×0.985 ≈ 0.913, and this value needs to be maximized.

[0110] Constraint Quantification: Battery Life Decay Rate: Limit the capacity decay within each control cycle ≤ 0.01% (corresponding to a 0.01% decrease in SOH); Grid Power Fluctuation: Limit |ΔP| ≤ 5% Setpoint (e.g., when Setpoint = 500kW, ΔP ≤ 25kW); PCS Hardware Limitations: Charge-Discharge Current ≤ Rated Value (e.g., 1000A), Switching Frequency ≤ 20kHz.

[0111] Multi-Objective Optimization Algorithm Implementation Algorithm Selection: Adopt the Non-Dominated Sorting Genetic Algorithm (NSGA-II) or Particle Swarm Optimization (PSO). For example, the NSGA-II is configured with population size = 50, crossover probability = 0.8, mutation probability = 0.1, and the number of iterations = 100 times.

[0112] Constraint Handling Strategy: Convert the constraints into penalty terms (Penalty Function). For example, if the capacity decay rate exceeds the standard, the objective function value is multiplied by a decay coefficient (e.g., 0.8); if the power fluctuation exceeds the standard, the objective function value is multiplied by 0.9.

[0113] Pareto Front Solving: Generate a non-dominated solution set (Pareto Front) through NSGA-II, such as 10 groups of solutions (combinations of charge-discharge current, switching frequency, and power factor compensation parameters) for the decision-making layer to select.

[0114] Solve the objective function using the rolling horizon optimization algorithm, iteratively calculate the optimal charge-discharge current value, PCS switching frequency, and power factor compensation parameters within each control cycle, and generate a dynamic PCS operating parameter candidate set; Rolling Horizon Optimization (RHC) Process Time-domain window division: Decompose the optimization problem into sub-problems executed in a rolling manner, where each sub-problem covers the next N control cycles (e.g., when N = 3, which is 15 seconds). For example, at the current time t = 0, optimize the parameters for t = 0 to 15 seconds, and at t = 5 seconds, roll and optimize the parameters for t = 5 to 20 seconds.

[0115] Dynamic parameter adjustment: Within each window, recalculate the objective function and constraint conditions based on the latest data. For example, at t = 0, the calculated current = 800A and the switching frequency = 15kHz; at t = 5 seconds, due to the increase in temperature, the current is reduced to 750A and the switching frequency is reduced to 12kHz.

[0116] Candidate set generation: Each set of solutions includes the charge and discharge current (I), switching frequency (f_sw), and power factor compensation angle (θ). For example: Candidate 1: I = 800A, f_sw = 15kHz, θ = 0.95 (capacitive compensation); Candidate 2: I = 780A, f_sw = 14kHz, θ = 0.98.

[0117] Real-time calculation and hardware acceleration Computing resource allocation: Use an FPGA (such as Xilinx Alveo U280) to accelerate the optimization algorithm, reducing the time taken for a single iteration from 10ms to 0.5ms.

[0118] Parallel computing optimization: Calculate multiple candidate solutions simultaneously (e.g., 50 sets of solutions), and select the optimal solution through priority sorting. For example, priority = objective function value × weight (efficiency weight = 0.6, lifespan weight = 0.4).

[0119] Parameter smooth transition: To avoid sudden parameter changes between adjacent cycles, impose a rate limit on parameters such as current and frequency (e.g., current change rate ≤ 50A / s). For example, the current current = 800A, and the maximum allowable value for the next cycle = 800 + 50×5 = 1050A.

[0120] Verify the feasibility of the candidate parameters in the candidate set of dynamic PCS operating parameters, detect whether the parameters exceed the safety boundary through the battery polarization voltage model and the grid harmonic analysis model, and output the instruction set for the dynamically adjusted PCS operating parameters that pass the verification.

[0121] Battery polarization voltage model verification Model construction: Based on an equivalent circuit model (such as a second-order RC model), calculate the polarization voltage (V_pol) of the battery at a specific current. For example, when the current I = 800A, V_pol = 0.05V (initial value) + 0.01V (temperature compensation).

[0122] Safety boundary detection: Limit the polarization voltage ≤ 0.1V (to prevent lithium plating). If the standard is exceeded, the candidate parameters are determined to be invalid. For example, for Candidate 1, V_pol = 0.12V, triggering the elimination mechanism.

[0123] Dynamic parameter correction: Linearly adjust the exceeded parameters. For example, reduce the current from 800A to 750A and recalculate V_pol = 0.09V, meeting the requirements.

[0124] Grid harmonic analysis and verification Harmonic distortion rate calculation: Analyze the harmonic components of the PCS output current through Fourier transform (FFT) and calculate the total harmonic distortion rate (THD). For example, when the switching frequency f_sw = 15kHz, THD = 4.5% (the standard requirement is ≤ 5%).

[0125] Filter parameter adjustment: If THD exceeds the standard (such as THD = 5.2%), increase the inductance value of the LC filter (such as from 100μH to 120μH) or adjust the switching frequency (such as from 15kHz to 16kHz).

[0126] Electromagnetic compatibility (EMC) test: Detect whether the high-frequency harmonics (such as 150kHz - 30MHz) exceed the radiation limit (such as the EN 55022 Class A standard). If the standard is exceeded, add magnetic rings or shielding layers.

[0127] Instruction set generation and distribution Verification passed instruction encapsulation: Package parameters such as current, frequency, and power factor into Modbus TCP instructions. For example: { "CMD": "SET_PCS_PARAM", "I": 750, "f_sw": 14, "theta": 0.98}.

[0128] Based on the collaborative optimization dataset, the model predictive control (MPC) framework uses a rolling time domain period of 5 - 15 minutes in the future to establish a multi-objective optimization function: with the battery life attenuation rate (such as capacity loss ≤ 0.01% / time) as a hard constraint and the grid power fluctuation (such as frequency deviation ≤ 0.2Hz) as a soft constraint, dynamically solve the optimal charge and discharge current, switching frequency, and power factor compensation parameters of the PCS. Through iterative optimization using the Lagrange multiplier method, generate dynamic control instructions that take into account both battery life and grid stability, break through the limitations of traditional single-objective control, achieve a dynamic balance between battery life and grid demand, and improve the economy and grid compatibility of the energy storage system.

[0129] S204. According to the PCS operating parameters, the internal temperature distribution of the high-voltage integrated machine is calculated in real time through a three-dimensional thermal field reconstruction model, the rotational speed of the cooling fan and the switching frequency of the power device are dynamically adjusted to generate a collaborative control instruction, and at the same time, it is fed back to the BMS and EMS to update the battery charge and discharge thresholds and the grid dispatching strategy, forming a closed-loop control link. Specifically, it may include: According to the PCS operating parameters, the power device loss value is calculated in real time by the finite element method, and a three-dimensional thermal field distribution model inside the high-voltage integrated machine is constructed by combining the fluid mechanics equation; ‌Power Device Loss Modeling and Finite Element Analysis‌ ‌Loss Decomposition and Parameter Acquisition‌: The losses of power devices (such as IGBT modules) are divided into conduction loss and switching loss. For example, when the PCS operating parameters are charge and discharge current I = 800A, switching frequency f_sw = 15kHz, and bus voltage V_dc = 1000V, the conduction loss is proportional to the square of the current (P_cond = I²×R_ds(on), when R_ds(on) = 0.5mΩ, P_cond = 800²×0.0005 = 320W), and the switching loss is related to the frequency and voltage (P_sw = 0.5×V_dc×I×t_sw×f_sw, when t_sw = 100ns, P_sw = 0.5×1000×800×100e - 9×15e3 = 600W). The total loss P_total = 320 + 600 = 920W.

[0130] ‌Finite Element Mesh Generation‌: Based on the three-dimensional geometric model of the high-voltage integrated machine (such as generated by SolidWorks or ANSYS SpaceClaim, as the three-dimensional thermal field reconstruction model), key components such as power devices, busbars, and heat dissipation fins are meshed, and the mesh size is set to 0.5mm to balance accuracy and calculation speed. For example, a certain IGBT module is divided into 500,000 hexahedral elements, and the radiator substrate is divided into 300,000 tetrahedral elements.

[0131] ‌Thermal-Electric Coupling Simulation‌: Using ANSYS Mechanical or COMSOL Multiphysics, the loss value is loaded as a heat source onto the mesh nodes to solve the steady-state temperature field. For example, the IGBT junction temperature T_junction = loss × thermal resistance (R_th = 0.1℃ / W) + ambient temperature (T_amb = 40℃), then T_junction = 920×0.1 + 40 = 132℃, and it needs to be further reduced to a safe value (such as <110℃) through fluid heat dissipation.

[0132] ‌Fluid Mechanics Equation and Heat Dissipation Channel Modeling‌ Parametric Modeling of Heat Dissipation Channels: According to the fan deployment plan (e.g., the number of axial fans = 4, the maximum air volume of a single fan Q = 500 CFM), a fluid domain model of the heat dissipation channels is constructed. The inlet boundary condition is set as the wind speed v_in = 5 m / s (corresponding to Q = 500 CFM), the outlet is a pressure outlet (P_out = 1 atm), and the surface of the heat dissipation fins is set as a no-slip wall surface.

[0133] Turbulence Model Selection: The k-ε turbulence model (applicable to high Reynolds number flows) or the SST k-ω model (applicable to near-wall flows) is adopted to solve the Navier-Stokes equations. For example, through Fluent calculation, the convective heat transfer coefficient h = 50 W / (m²·K) on the surface of the heat dissipation fins is obtained, and the fin temperature drops from 132 °C to 98 °C.

[0134] Coupled Iteration of Thermal Field and Flow Field: The temperature field calculated by finite element is used as the input for fluid simulation to update physical property parameters such as air density and viscosity, and iterated until the residual converges (e.g., the residual of the energy equation < 1e-6). Finally, a three-dimensional temperature distribution cloud map inside the high-voltage integrated machine is output as a three-dimensional thermal field distribution model, marking the key hot spot positions (e.g., the center temperature of the IGBT module = 112 °C, which needs further optimization).

[0135] Real-time Calculation and Hardware Acceleration GPU-accelerated Solving: The NVIDIA A100 GPU is used for parallel calculation of finite element and fluid mechanics equations, and the single simulation time is shortened from the minute level to the second level (e.g., a complete thermal field update is completed in 10 seconds).

[0136] Application of Reduced Order Model (ROM): For working conditions with high repeatability (such as constant current charge and discharge), the thermal field characteristic modes are extracted by principal component analysis (PCA) to construct a ROM to replace the full-order model, and the calculation speed is increased to the millisecond level (e.g., the temperature change is predicted within 1 ms).

[0137] Online Calibration Mechanism: The actual temperature data is collected in real time through an infrared thermal imager (such as FLIR A655sc) and compared with the simulation results. If the deviation exceeds ±5 °C, the model parameter adaptive correction is triggered (such as adjusting the convective heat transfer coefficient or the material thermal conductivity).

[0138] Identify the temperature-exceeding areas in the three-dimensional thermal field distribution model, dynamically adjust the speed gears of the heat dissipation fans and the switching frequencies of the power devices, and generate a draft heat dissipation control instruction; Temperature-exceeding Area Detection and Priority Ranking Threshold determination and area marking: Define the temperature safety threshold (e.g., T_safe = 110 °C), and mark all areas where the temperature exceeds the threshold by traversing the nodes of the three-dimensional thermal field grid (e.g., the temperatures of IGBT modules A1, A2, and B1 are 112 °C, 115 °C, and 108 °C respectively).

[0139] Thermal hazard level classification: Classify according to the over-temperature range and component importance. For example: First-level over-temperature (T > 120 °C): Immediately reduce the load or shut down the machine; Second-level over-temperature (115 °C < T ≤ 120 °C): Forcefully increase the heat dissipation intensity; Third-level over-temperature (110 °C < T ≤ 115 °C): Locally adjust the fan and switching frequency.

[0140] Dynamic priority allocation: Use the weighted scoring method, score = over-temperature range × weight (e.g., IGBT weight = 0.7, capacitor weight = 0.3), and give priority to processing the area with the highest score. For example, the score of IGBT module A2 = (115 - 110) × 0.7 = 3.5, which needs to be adjusted first.

[0141] Coordinated control of the cooling fan and switching frequency Fan speed adjustment strategy: Match the corresponding fan according to the location of the over-temperature area. For example, if IGBT module A1 is cooled by fan FAN1, increase its speed from 2000 RPM to 2500 RPM (corresponding air volume increases from 400 CFM to 500 CFM), and at the same time limit the maximum speed not to exceed 3000 RPM (to avoid excessive noise).

[0142] Dynamic adjustment of the switching frequency: Reduce the switching frequency of the power device in the over-temperature area to reduce losses. For example, reduce the switching frequency of IGBT module A2 from 15 kHz to 12 kHz, so that the switching loss is reduced from 600 W to 480 W (a 20% reduction), and the junction temperature is reduced from 115 °C to 109 °C.

[0143] Multivariable coupling optimization: Use a fuzzy logic controller (FLC) to coordinate the fan speed and switching frequency. The input variables are the temperature deviation (ΔT = T - T_safe) and its change rate (dΔT / dt), and the output variables are the fan speed increment (ΔRPM) and the switching frequency adjustment amount (Δf_sw). For example, the rule base defines: If ΔT = 5 °C and dΔT / dt = +1 °C / s, then ΔRPM = +500, Δf_sw = -3 kHz; If ΔT = 3 °C and dΔT / dt = -0.5 °C / s, then ΔRPM = +300, Δf_sw = -1 kHz.

[0144] Draft instruction generation and conflict resolution Instruction conflict detection: If there are contradictions among multiple adjustment instructions (for example, increasing the speed of a certain fan may cause overcooling in adjacent areas), a compromise solution is found through a game theory model (such as Nash equilibrium). For example, increasing the speed of fan FAN1 to 2500 RPM will reduce the temperature of capacitor C1 from 95°C to 88°C, but may increase system noise. The optimal solution needs to be selected after weighing.

[0145] Instruction encoding and encapsulation: Encode the adjustment parameters into a JSON format instruction draft. For example: { "cmd_id": "THERMAL_CTRL_001", "timestamp": "2023-10-01T14:30:00Z", "actions": {"target": "FAN1", "param": "rpm", "value": 2500}, {"target": "IGBT_A2", "param": "f_sw", "value": 12} }。

[0146] Risk assessment and alternative solutions: Conduct a failure mode and effects analysis (FMEA) on the instruction draft. For example, if fan FAN1 fails and cannot increase speed, an alternative solution is enabled (such as turning off secondary loads to reduce total power consumption).

[0147] Input the heat dissipation control instruction draft into the digital twin system for simulation execution, predict the temperature change trend after adjustment. If the simulation result does not meet the expectation, trigger instruction replanning until the thermal management requirements are met, and obtain the final coordinated control instruction; Digital twin system modeling and real-time synchronization Twin model construction: Create a digital twin of the high-voltage integrated machine based on Unity3D or Siemens NX, and map the three-dimensional thermal field, electrical parameters, and mechanical state of the physical entity in real time. For example, parameters such as the temperature of the IGBT module, fan speed, and switching frequency are synchronized once per second.

[0148] Simulation engine configuration: Integrate Modelica or Simulink as the background simulation engine, and define the coupled equations of thermal-electric-fluid multi-physics fields. For example, set the simulation step size Δt = 1 second to predict the temperature change in the next 60 seconds.

[0149] ​Instruction Injection and Scenario Triggering: Inject a draft instruction (such as increasing the speed of FAN1 to 2500 RPM) into the digital twin to activate a preset simulation scenario (such as "Thermal Management Test under Extreme FM Load").

[0150] Predictive Simulation and Result Evaluation Short-term Trend Prediction: Run the simulation for 5 - 10 seconds and output key metrics (such as IGBT junction temperature, radiator outlet air velocity, total system power consumption). For example, the prediction results show that after increasing the speed of FAN1, the temperature of IGBT_A2 drops from 115°C to 108°C within 5 seconds, but the power consumption of FAN1 increases from 200W to 250W.

[0151] Multi-objective Compliance Judgment: Check whether the prediction results meet all constraint conditions: Temperature Constraint: T ≤ 110°C; Noise Constraint: Fan noise ≤ 75 dB(A); Efficiency Constraint: System efficiency ≥ 90%.

[0152] If any condition is not met (such as noise = 78 dB(A)), the draft instruction is determined to be invalid.

[0153] Sensitivity Analysis: Evaluate the impact of parameter uncertainty through Monte Carlo simulation. For example, if the wind speed fluctuates by ±10%, the predicted temperature may fluctuate by ±2°C, and it is necessary to ensure that the temperature does not exceed the limit in the worst case.

[0154] Instruction Reprogramming and Iterative Optimization Root Cause Analysis and Parameter Adjustment: If the prediction results do not meet the standards, locate the cause of failure and adjust the instruction. For example, if the noise exceeds the standard due to the increase in the fan speed, the switching frequency can be reduced (from 15 kHz → 10 kHz) and the rotational speed can be slightly increased (2000 → 2200 RPM) at the same time.

[0155] Parallel Testing of Multiple Schemes: Run multiple candidate draft instructions simultaneously in the digital twin (such as Scheme A: frequency reduction + medium-speed fan; Scheme B: turn off the secondary circuit + high-speed fan), and select the scheme with the highest comprehensive score. Example of the scoring formula: Score = 0.6×(temperature drop) + 0.2×(efficiency loss) + 0.2×(noise reduction).

[0156] Convergence Condition and Termination Mechanism: Set the maximum number of iterations (such as 10 times) or the error tolerance (such as temperature deviation ≤ 1°C), and output the final instruction when the condition is met. For example, after 3 iterations, a feasible scheme is obtained: FAN1 = 2300 RPM, f_sw of IGBT_A2 = 11 kHz, predicted temperature = 109°C, noise = 73 dB(A).

[0157] Send the final coordinated control instruction to the PCS and the heat dissipation system for execution. At the same time, feedback the temperature distribution data to the BMS to update the battery charge and discharge threshold, and send a correction suggestion for the grid dispatching strategy to the EMS to form a closed-loop control link.

[0158] Instruction Sending and Equipment Control Protocol and Interface Adaptation: Send instructions to the PCS and the fan controller through the CAN bus or the EtherCAT protocol. For example, send CAN frame data: ID: 0x301, Data: [0x12 0x34 0x56] / / 0x12 = Fan FAN1, 0x34 = Rotation speed 2300 RPM; ID: 0x302, Data: [0xA2 0x0B 0x00] / / 0xA2 = IGBT_A2, 0x0B = Switching frequency 11kHz.

[0159] Execution Status Monitoring: Read the equipment feedback in real time (such as the actual rotation speed of the fan, the IGBT temperature). If the deviation between the actual parameter and the instruction exceeds ±5% (such as the fan speed error > 100 RPM), trigger an abnormal alarm and start redundant control (such as switching to the standby fan).

[0160] Timing Synchronization and Queuing: Use a priority queue to manage the instruction execution order. For example, the priority of the temperature control instruction = high, the priority of the power factor compensation instruction = medium, to ensure that emergency operations are executed first.

[0161] Data Feedback and Threshold Update Dynamic Adjustment of the BMS Charge and Discharge Threshold: Calculate the maximum allowable charge and discharge current of the battery according to the temperature distribution data (such as the highest temperature = 109°C). For example, the original threshold I_max = 1000A, according to the temperature-current derating curve (derating 5% for every 10°C increase), adjust to I_max = 1000×(1-(109 - 25) / 10×0.05)=1000×0.58 = 580A.

[0162] State of Health (SOH) Compensation: If the battery SOH < 80%, further reduce the threshold (such as I_max = 580×0.8 = 464A), and update the parameters through the BMS's UDS protocol (ISO 14229).

[0163] Historical Data Storage and Analysis: Store data such as temperature, current, and SOH in a time series database (such as InfluxDB) for subsequent life prediction and maintenance decision-making.

[0164] Grid Dispatching Strategy Collaborative Optimization EMS Strategy Revision Suggestion Generation: Send scheduling suggestions to the EMS based on heat dissipation capacity limitations (e.g., current maximum sustainable power = 500 kW). For example: { "type": "POWER_LIMIT_ADJUST", "new_max_power": 500, "duration": 300, "reason": "thermal_constraint" }。

[0165] Demand Response (DR) Participation Adjustment: If the power grid needs to call the energy storage system during peak hours but the system's heat dissipation has reached its limit, it is recommended that the EMS reduce the frequency modulation participation rate (e.g., from 100% → 70%) or apply for a delayed response (e.g., execute after 10 minutes).

[0166] Closed-loop Link Self-verification: Evaluate the closed-loop control effect by comparing key indicators (such as temperature, efficiency, power grid response error) before and after the execution of the instructions, and automatically optimize control parameters (such as PID gain, fuzzy rule base).

[0167] Technical Details and Parameter Examples Finite Element Simulation Parameters Number of meshes: 500,000 cells for the IGBT module and 300,000 cells for the radiator; Material properties: Thermal conductivity of the copper busbar is 401 W / (m·K), and that of the aluminum radiator is 237 W / (m·K); Boundary conditions: Ambient temperature is 40°C, and the initial wind speed is 5 m / s.

[0168] Digital Twin Simulation Configuration Simulation step size: 1 second; Prediction duration: 60 seconds; Number of Monte Carlo samples: 1000 times.

[0169] Control Instruction Parameters Fan speed range: 1000 - 3000 RPM; Switching frequency range: 5 - 20 kHz; Temperature sampling frequency: 10 Hz.

[0170] Application Scenario Examples Scenario Description: When a certain energy storage power station participates in power grid frequency modulation, the ambient temperature suddenly rises (40°C → 45°C), resulting in overheating of the IGBT module.

[0171] Thermal field modeling: FEM calculates the IGBT loss of 920W and predicts the junction temperature of 132°C; Instruction generation: Increase the fan FAN1 to 2500 RPM and reduce the switching frequency to 12kHz; Digital twin verification: The predicted temperature drops to 109°C, but the noise = 76dB(A) exceeds the standard; Instruction re-planning: Adjust to fan 2300 RPM, switching frequency 11kHz, temperature = 110°C, noise = 73dB(A); Closed-loop execution: Update the BMS charge and discharge threshold to 580A and recommend the EMS to limit the power to 500kW.

[0172] Based on the PCS operating parameters (such as IGBT switching frequency, current value), the loss value of the power device is calculated in real time by the finite element method, and a three-dimensional thermal field distribution model (resolution up to 1mm³) is constructed by combining the wind speed data of the heat dissipation channel. After identifying the area with excessive temperature (such as >85°C), the fan speed (such as rising from 2000rpm to 3500rpm) and switching frequency (such as dropping from 10kHz to 8kHz) are dynamically adjusted, and the control effect is verified by the digital twin, and finally the collaborative control instruction is generated. At the same time, the temperature data is fed back to the BMS to adjust the charge and discharge threshold (such as limiting the current by 10%), and a grid dispatching correction suggestion (such as reducing the frequency modulation response speed) is sent to the EMS to realize the closed-loop collaboration of the thermal-electric-control multi-physical fields and avoid equipment failures caused by local overheating; the system adaptive ability is improved through the real-time feedback mechanism to ensure the safety and response accuracy of the energy storage system under complex working conditions.

[0173] It can be seen that according to the power level and thermal distribution requirements of the energy storage system, the high-voltage box and the power conversion system PCS are designed for three-dimensional space coupling through the topology optimization algorithm to generate the physical layer architecture scheme of the high-voltage integrated machine; based on the physical layer architecture scheme of the high-voltage integrated machine, the collaborative optimization data set is constructed by using the timestamp alignment technology; the collaborative optimization data set is input into the model predictive control framework to generate the dynamically adjusted PCS operating parameters; according to the PCS operating parameters, the internal temperature distribution of the high-voltage integrated machine is calculated in real time by the three-dimensional thermal field reconstruction model, and the fan speed of the heat dissipation and the switching frequency of the power device are dynamically adjusted to generate the collaborative control instruction, so as to improve the integration level and collaborative control ability of the high-voltage energy storage system and realize the efficient and stable operation of the energy storage system.

[0174] Another embodiment of the present invention provides an energy storage system high-voltage integration and collaborative control system, see Figure 3 The system may include: The design module 301 is configured to perform a three-dimensional spatial coupling design of the high-voltage box and the power conversion system PCS according to the power level and thermal distribution requirements of the energy storage system through a topology optimization algorithm, and generate a physical layer architecture solution for the high-voltage integrated machine. Among them, the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel; The construction module 302 is configured to synchronously collect the real-time battery state data of the battery management system BMS and the grid dispatching instructions of the energy management system EMS based on the physical layer architecture solution of the high-voltage integrated machine through a dynamic data bus, and construct a collaborative optimization data set by using the timestamp alignment technology. Among them, the collaborative optimization data set includes multi-dimensional coupling characteristics such as battery health, charge and discharge rate, and grid demand response; The generation module 303 is configured to input the collaborative optimization data set into the model predictive control framework, solve the charge and discharge strategy by using the rolling horizon optimization algorithm, and generate dynamically adjusted PCS operating parameters. Among them, the rolling horizon optimization algorithm realizes multi-objective collaborative optimization by restricting the battery life attenuation rate and the grid power fluctuation range; The control module 304 is configured to calculate the internal temperature distribution of the high-voltage integrated machine in real time through a three-dimensional thermal field reconstruction model according to the PCS operating parameters, dynamically adjust the rotation speed of the cooling fan and the switching frequency of the power device, generate a collaborative control instruction, and at the same time feedback to the BMS and EMS to update the battery charge and discharge threshold and the grid dispatching strategy to form a closed-loop control link.

[0175] It can be seen that according to the power level and thermal distribution requirements of the energy storage system, a three-dimensional spatial coupling design of the high-voltage box and the power conversion system PCS is performed through a topology optimization algorithm to generate a physical layer architecture solution for the high-voltage integrated machine; based on the physical layer architecture solution of the high-voltage integrated machine, a collaborative optimization data set is constructed by using the timestamp alignment technology; the collaborative optimization data set is input into the model predictive control framework to generate dynamically adjusted PCS operating parameters; according to the PCS operating parameters, the internal temperature distribution of the high-voltage integrated machine is calculated in real time through a three-dimensional thermal field reconstruction model, and the rotation speed of the cooling fan and the switching frequency of the power device are dynamically adjusted to generate a collaborative control instruction, so as to improve the integration level and collaborative control ability of the high-voltage energy storage system and realize the efficient and stable operation of the energy storage system.

[0176] An embodiment of the present invention further provides a storage medium, in which a computer program is stored. Among them, the computer program is set to execute the steps in any one of the above method embodiments when running.

[0177] Specifically, in this embodiment, the above storage medium can be set to store a computer program for executing the following steps: S201. According to the power level and thermal distribution requirements of the energy storage system, the high-voltage box and the power conversion system PCS are designed for three-dimensional space coupling through a topology optimization algorithm to generate a physical layer architecture solution for the high-voltage integrated machine. Among them, the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel; S202. Based on the physical layer architecture solution of the high-voltage integrated machine, the real-time battery state data of the battery management system BMS and the grid dispatching instructions of the energy management system EMS are synchronously collected through a dynamic data bus, and a collaborative optimization data set is constructed using the timestamp alignment technology. Among them, the collaborative optimization data set contains multi-dimensional coupling characteristics such as battery health, charge and discharge rate, and grid demand response; S203. Input the collaborative optimization data set into the model predictive control framework, and use the rolling horizon optimization algorithm to solve the charge and discharge strategy to generate dynamically adjusted PCS operating parameters. Among them, the rolling horizon optimization algorithm realizes multi-objective collaborative optimization by constraining the battery life attenuation rate and the grid power fluctuation range; S204. According to the PCS operating parameters, the internal temperature distribution of the high-voltage integrated machine is calculated in real time through a three-dimensional thermal field reconstruction model, the rotation speed of the cooling fan and the switching frequency of the power device are dynamically adjusted to generate a collaborative control instruction, and at the same time, it is fed back to the BMS and EMS to update the battery charge and discharge threshold and the grid dispatching strategy to form a closed-loop control link.

[0178] It can be seen that according to the power level and thermal distribution requirements of the energy storage system, the high-voltage box and the power conversion system PCS are designed for three-dimensional space coupling through a topology optimization algorithm to generate a physical layer architecture solution for the high-voltage integrated machine; based on the physical layer architecture solution of the high-voltage integrated machine, a collaborative optimization data set is constructed using the timestamp alignment technology; the collaborative optimization data set is input into the model predictive control framework to generate dynamically adjusted PCS operating parameters; according to the PCS operating parameters, the internal temperature distribution of the high-voltage integrated machine is calculated in real time through a three-dimensional thermal field reconstruction model, the rotation speed of the cooling fan and the switching frequency of the power device are dynamically adjusted to generate a collaborative control instruction, so as to improve the integration level and collaborative control ability of the high-voltage energy storage system and realize the efficient and stable operation of the energy storage system.

[0179] An embodiment of the present invention further provides an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0180] Specifically, the above electronic device may further include a transmission device and an input / output device. Among them, the transmission device is connected to the above processor, and the input / output device is connected to the above processor.

[0181] Specifically, in this embodiment, the above-mentioned processor can be set to execute the following steps through a computer program: S201. According to the power level and thermal distribution requirements of the energy storage system, perform three-dimensional space coupling design on the high-voltage box and the power conversion system PCS through a topology optimization algorithm to generate a physical layer architecture scheme of the high-voltage integrated machine, wherein the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel; S202. Based on the physical layer architecture scheme of the high-voltage integrated machine, synchronously collect the real-time battery state data of the battery management system BMS and the grid dispatching instructions of the energy management system EMS through a dynamic data bus, and construct a collaborative optimization data set by using the timestamp alignment technology, wherein the collaborative optimization data set includes multi-dimensional coupling characteristics such as battery health, charge and discharge rate, and grid demand response; S203. Input the collaborative optimization data set into the model predictive control framework, and use the rolling horizon optimization algorithm to solve the charge and discharge strategy to generate dynamically adjusted PCS operating parameters, wherein the rolling horizon optimization algorithm realizes multi-objective collaborative optimization by restricting the battery life attenuation rate and the grid power fluctuation range; S204. According to the PCS operating parameters, calculate the internal temperature distribution of the high-voltage integrated machine in real time through a three-dimensional thermal field reconstruction model, dynamically adjust the rotation speed of the cooling fan and the switching frequency of the power device, generate a collaborative control instruction, and at the same time feedback to the BMS and EMS to update the battery charge and discharge threshold and the grid dispatching strategy to form a closed-loop control link.

[0182] It can be seen that according to the power level and thermal distribution requirements of the energy storage system, perform three-dimensional space coupling design on the high-voltage box and the power conversion system PCS through a topology optimization algorithm to generate a physical layer architecture scheme of the high-voltage integrated machine; based on the physical layer architecture scheme of the high-voltage integrated machine, construct a collaborative optimization data set by using the timestamp alignment technology; input the collaborative optimization data set into the model predictive control framework to generate dynamically adjusted PCS operating parameters; according to the PCS operating parameters, calculate the internal temperature distribution of the high-voltage integrated machine in real time through a three-dimensional thermal field reconstruction model, dynamically adjust the rotation speed of the cooling fan and the switching frequency of the power device, generate a collaborative control instruction, so as to improve the integration level and collaborative control ability of the high-voltage energy storage system and realize the efficient and stable operation of the energy storage system.

[0183] The above has detailed the structure, characteristics and function effects of the present invention according to the embodiments shown in the drawings. The above is only the preferred embodiment of the present invention, but the present invention is not limited to the scope shown in the drawings. Any changes made according to the concept of the present invention, or modified into equivalent embodiments with equivalent changes, still within the spirit covered by the specification and the drawings, shall be within the protection scope of the present invention.

Claims

1. A high-voltage integration and coordinated control method for an energy storage system, characterized in that The method includes: According to the power level and thermal distribution requirements of the energy storage system, the high-voltage box and the power conversion system PCS are designed for three-dimensional space coupling through a topology optimization algorithm to generate a physical layer architecture scheme of the integrated high-voltage unit. Among them, the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel; Based on the physical layer architecture scheme of the integrated high-voltage unit, the real-time battery state data of the battery management system BMS and the grid dispatching instructions of the energy management system EMS are synchronously collected through a dynamic data bus, and a collaborative optimization data set is constructed by using the timestamp alignment technology. Among them, the collaborative optimization data set contains multi-dimensional coupling characteristics such as battery health, charge and discharge rate, and grid demand response; The collaborative optimization data set is input into the model predictive control framework, and a rolling horizon optimization algorithm is used to solve the charge and discharge strategy to generate dynamically adjusted PCS operating parameters. Among them, the rolling horizon optimization algorithm realizes multi-objective collaborative optimization by restricting the battery life attenuation rate and the grid power fluctuation range; According to the PCS operating parameters, the internal temperature distribution of the integrated high-voltage unit is calculated in real time through a three-dimensional thermal field reconstruction model, the rotation speed of the cooling fan and the switching frequency of the power device are dynamically adjusted to generate collaborative control instructions, and at the same time, feedback is sent to the BMS and EMS to update the battery charge and discharge thresholds and the grid dispatching strategy to form a closed-loop control link.

2. The method according to claim 1, wherein The step of designing the three-dimensional space coupling of the high-voltage box and the power conversion system PCS through a topology optimization algorithm according to the power level and thermal distribution requirements of the energy storage system to generate a physical layer architecture scheme of the integrated high-voltage unit, where the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel, includes: According to the power level parameters and thermal distribution requirements of the energy storage system, a three-dimensional initial model of the high-voltage box and PCS is constructed to generate an initial three-dimensional grid model including electrical interfaces, heat dissipation port positions, and internal cavities; In the initial three-dimensional grid model, the electrical connection path is designed based on the minimum path principle, and the length and impedance of the busbar wiring are iteratively optimized through the ant colony algorithm to generate a candidate set of electrical connection paths and mark the key node coordinates; At the same time, a hydrodynamic simulation of the heat dissipation channel is carried out, the layout of the heat dissipation fins is optimized according to the heat source distribution density, and the heat dissipation efficiency curve at different wind speeds is calculated to generate the topological structure of the heat dissipation channel and the fan deployment scheme; The candidate set of electrical connection paths and the topological structure of the heat dissipation channel are input into a multi-objective optimization algorithm, and the Pareto optimal solution set is solved with the electrical loss rate and the heat dissipation efficiency as the constraint conditions to generate an electrical-heat dissipation coupled topological scheme; The electromagnetic-thermal coupling simulation verification is carried out on the topological scheme, the local hot spots and the areas with excessive electromagnetic interference are detected, the connection path and the heat dissipation channel spacing are corrected, and the three-dimensional scheme of the physical layer architecture of the final integrated high-voltage unit is output.

3. The method according to claim 2, wherein Based on the physical layer architecture of the high-voltage integrated machine, real-time battery status data of the battery management system (BMS) and grid dispatching instructions of the energy management system (EMS) are synchronously collected through a dynamic data bus, and a collaborative optimization data set is constructed using timestamp alignment technology. The collaborative optimization data set includes multi-dimensional coupling characteristics of battery health, charge and discharge rate, and grid demand response, including: Collect battery status data of the BMS and grid dispatching instructions of the EMS through the dynamic data bus, and use hardware clock synchronization technology to align data timestamps to generate a time-synchronized original data stream; Clean outliers from the original data stream, compensate for missing data based on the battery cluster topology relationship, and use the covariance matrix to eliminate sensor noise interference to generate a purified multi-dimensional monitoring data set; Extract battery health characteristics, charge and discharge rate characteristics, and grid demand response characteristics from the multi-dimensional monitoring data set to construct a multi-dimensional coupling characteristic correlation matrix; According to the multi-dimensional coupling characteristic correlation matrix, analyze the dynamic correlation between battery health, charge and discharge rate, and grid demand response pairwise through the Pearson correlation coefficient to generate a multi-dimensional coupling characteristic mapping table containing weight labels; Fuse the multi-dimensional coupling characteristic mapping table with time series data, and package it into a collaborative optimization data set with spatio-temporal labels according to a preset sampling period, and store it in the shared memory area of the edge computing node.

4. The method according to claim 3, wherein Input the collaborative optimization data set into the model predictive control framework, and use the rolling horizon optimization algorithm to solve the charge and discharge strategy to generate dynamically adjusted PCS operating parameters. The rolling horizon optimization algorithm realizes multi-objective collaborative optimization by restricting the battery life attenuation rate and the grid power fluctuation range, including: Input the collaborative optimization data set into the model predictive control framework, predict the power demand curve for a preset number of future control cycles based on the current battery status and grid instructions, and generate a reference trajectory for the charge and discharge power benchmark; Establish a multi-objective optimization model, refer to the reference trajectory of the charge and discharge power benchmark, use the battery life attenuation rate as the first constraint and the grid power fluctuation range as the second constraint, and define the objective function as maximizing the system efficiency; Use the rolling horizon optimization algorithm to solve the objective function, iteratively calculate the optimal charge and discharge current values, PCS switching frequency, and power factor compensation parameters in each control cycle to generate a candidate set of dynamic PCS operating parameters; Verify the feasibility of the candidate parameters in the candidate set of dynamic PCS operating parameters, and detect whether the parameters exceed the safety boundary through the battery polarization voltage model and the grid harmonic analysis model, and output a set of dynamically adjusted PCS operating parameter instructions that pass the verification.

5. The method according to claim 4, wherein According to the PCS operating parameters, calculate the internal temperature distribution of the high-voltage integrated machine in real time through a three-dimensional thermal field reconstruction model, dynamically adjust the rotation speed of the cooling fan and the switching frequency of the power device to generate collaborative control instructions, and at the same time feedback to the BMS and EMS to update the battery charge and discharge threshold and the grid dispatching strategy to form a closed-loop control link, including: According to the PCS operating parameters, calculate the power device loss value in real time through the finite element method, and combine the hydrodynamics equation to construct a three-dimensional thermal field distribution model inside the high-voltage integrated machine; Identify the temperature over - standard area in the three - dimensional thermal field distribution model, dynamically adjust the rotation speed gear of the cooling fan and the switching frequency of the power device, and generate a draft of the heat dissipation control instruction; Input the draft of the heat dissipation control instruction into the digital twin system for simulation execution, predict the temperature change trend after adjustment. If the simulation result does not meet the expectation, trigger the instruction replanning until the thermal management requirements are met, and obtain the final coordinated control instruction; Send the final coordinated control instruction to the PCS and the heat dissipation system for execution. At the same time, feedback the temperature distribution data to the BMS to update the battery charge - discharge threshold, and send a correction suggestion for the grid dispatching strategy to the EMS to form a closed - loop control link.

6. A high-voltage integrated and coordinated control system for an energy storage system, characterized in that, The system includes: A design module, which is used to couple the high - voltage box and the power conversion system PCS in three - dimensional space according to the power level and thermal distribution requirements of the energy storage system through a topology optimization algorithm, and generate a physical layer architecture scheme of the high - voltage integrated machine. Among them, the topology optimization algorithm synchronously calculates the optimal matching relationship between the electrical connection path and the heat dissipation channel; A construction module, which is used to synchronously collect the real - time battery state data of the battery management system BMS and the grid dispatching instruction of the energy management system EMS based on the physical layer architecture scheme of the high - voltage integrated machine through a dynamic data bus, and construct a collaborative optimization data set by using the timestamp alignment technology. Among them, the collaborative optimization data set contains multi - dimensional coupling characteristics such as battery health, charge - discharge rate, and grid demand response; A generation module, which is used to input the collaborative optimization data set into the model predictive control framework, solve the charge - discharge strategy by using the rolling - horizon optimization algorithm, and generate dynamically adjusted PCS operating parameters. Among them, the rolling - horizon optimization algorithm realizes multi - objective collaborative optimization by constraining the battery life attenuation rate and the grid power fluctuation range; A control module, which is used to calculate the internal temperature distribution of the high - voltage integrated machine in real - time according to the PCS operating parameters through a three - dimensional thermal field reconstruction model, dynamically adjust the rotation speed of the cooling fan and the switching frequency of the power device, generate a coordinated control instruction, and at the same time feedback to the BMS and the EMS to update the battery charge - discharge threshold and the grid dispatching strategy to form a closed - loop control link.

7. The system according to claim 6, characterized in that, The design module is specifically used for: According to the power level parameters and thermal distribution requirements of the energy storage system, construct a three - dimensional initial model of the high - voltage box and the PCS, and generate an initial three - dimensional grid model including electrical interfaces, heat dissipation port positions, and internal cavities; In the initial three - dimensional grid model, design the electrical connection path based on the minimum path principle, iteratively optimize the busbar wiring length and impedance through the ant colony algorithm, generate a candidate set of electrical connection paths and mark the key node coordinates; Synchronously perform a hydrodynamic simulation on the heat dissipation channel, optimize the layout of the heat dissipation fins according to the heat source distribution density, calculate the heat dissipation efficiency curve at different wind speeds, and generate the topological structure of the heat dissipation channel and the fan deployment plan; Input the candidate set of electrical connection paths and the topological structure of the heat dissipation channel into the multi - objective optimization algorithm, take the electrical loss rate and the heat dissipation efficiency as constraint conditions, solve the Pareto optimal solution set, and generate an electrical - heat dissipation coupled topological scheme; Conduct electromagnetic-thermal coupling simulation verification on the topology scheme, detect local hot spots and areas with excessive electromagnetic interference, correct the connection path and the spacing of the heat dissipation channels, and output the three-dimensional physical layer architecture scheme of the final high-voltage integrated machine.

8. The system according to claim 7, wherein The building block is specifically used for: Collect the battery status data of the BMS and the grid dispatching instructions of the EMS through the dynamic data bus, use the hardware clock synchronization technology to align the data timestamps, and generate the original data stream with time synchronization. Clean the outliers of the original data stream, compensate for the missing data based on the battery cluster topology relationship, and use the covariance matrix to eliminate the sensor noise interference to generate a purified multi-dimensional monitoring data set. Extract the battery health characteristics, charge and discharge rate characteristics, and grid demand response characteristics from the multi-dimensional monitoring data set to construct a multi-dimensional coupling feature correlation matrix. According to the multi-dimensional coupling feature correlation matrix, analyze the dynamic correlation relationship between the battery health, charge and discharge rate, and grid demand response pairwise through the Pearson correlation coefficient, and generate a multi-dimensional coupling feature mapping table containing weight labels. Fuse the multi-dimensional coupling feature mapping table with the time series data, package it into a collaborative optimization data set with spatio-temporal labels according to the preset sampling period, and store it in the shared memory area of the edge computing node.

9. A storage medium, characterized in that, A computer program is stored in the storage medium, wherein the computer program is set to execute the method according to any one of claims 1-5 when running.

10. An electronic device, comprising a memory and a processor, characterized in that, A computer program is stored in the memory, and the processor is set to run the computer program to execute the method according to any one of claims 1-5.

Images