Source network load storage cooperative control electromagnetic transient real-time simulation system and method

Abstract

The invention relates to a source network load storage cooperative control electromagnetic transient real-time simulation system and method. The method comprises the following steps: a data acquisition and preprocessing module performs data acquisition and preprocessing; the electromagnetic transient simulation modeling module performs electromagnetic transient simulation model construction and dynamic adaptive optimization; the multi-dimensional simulation precision verification module performs multi-dimensional precision verification on the simulation model, and the semi-physical real-time simulation system constructs a reliable communication link between the simulation calculation server and the energy management controller; and the closed-loop simulation verification module realizes closed-loop linkage of the simulation model and the energy management controller, and verifies the effectiveness of the control strategy. According to the method, the electric quantity data is directly intercepted from the memory of the simulation calculation server, the dependence on a physical interface box or a special server is avoided, actual engineering factors are brought into a simulation closed loop, the real performance that the controller adjusts the wind / light / storage output to stabilize and support under the working conditions of load and new energy fluctuation, system faults and the like is effectively verified, and the reliability of the simulation calculation server is improved. And a basis is provided for field parameter setting.

Description

Technical Field

[0001] This application relates to the field of integrated source-grid-load-storage system control and simulation test technology, specifically to a source-grid-load-storage collaborative control electromagnetic transient real-time simulation system and method. Background Technology

[0002] Against the backdrop of carbon reduction demands, many regions are promoting the construction of integrated power generation, grid, load, and energy storage projects in large-scale, energy-intensive industrial parks (such as silicon production, iron smelting, and chemical industries). These projects all require the construction of a smart control system. A coordinated control strategy at the power generation, grid, load, and energy storage system level is one of the key technologies of such a system. Due to the significant differences in load characteristics among different production loads, customized control strategies are needed. This involves coordinating the control of wind power, photovoltaic power, and energy storage systems to ensure the stable operation of production loads and maximize the utilization of new energy sources.

[0003] However, existing source-grid-load-storage control strategy verification technologies suffer from the following problems: 1. High cost due to reliance on hardware interfaces: Traditional hardware-in-the-loop (HIL) testing requires physical interface boxes or I / O boards to convert the digital signals output by the simulator into analog voltage / current signals before transmitting them to the controller. This not only increases hardware costs but also introduces complex wiring work and signal attenuation / interference issues. Traditional pure software offline simulations also ignore the time consumption and transmission delay of communication protocol parsing in actual engineering projects. 2. Distortion in simulation of high-energy-consuming load characteristics: Existing simulations often focus on the power supply side, typically directly calling the integrated load module built into the simulation software to simplify the load into a constant power or constant impedance model. However, actual industrial loads have extremely strong nonlinearity, impulsiveness, and harmonic characteristics. Simple general models cannot reflect the instantaneous impact of load fluctuations on grid voltage and frequency, causing the control strategy to lag or oscillate during the "source follows load" adjustment process. 3. Difficulty in integrating heterogeneous models: Actual engineering projects involve black-box models (wind / solar / storage) of equipment from multiple manufacturers and custom load models, lacking a unified hardware-in-the-loop integration verification environment. There is an urgent need for a hardware-in-the-loop simulation system that can eliminate physical interfaces, simulate high-energy-consuming load changes as accurately as possible, and verify the control strategies of real controllers.

[0004] Therefore, this invention proposes a source-grid-load-storage coordinated control electromagnetic transient real-time simulation system and method. Summary of the Invention

[0005] To overcome the shortcomings of the prior art, this application provides a source-grid-load-storage coordinated control electromagnetic transient real-time simulation system, specifically adopting the following technical solution.

[0006] A real-time simulation system for electromagnetic transients of source-grid-load-storage coordinated control includes:

[0007] Data Acquisition and Preprocessing Module: Enables accurate acquisition, standardized preprocessing, and secure and reliable storage of data across the entire power grid, source, load, and storage system. It includes a data acquisition unit, a data preprocessing unit, and a data storage unit.

[0008] Data acquisition unit: Real-time acquisition of multi-dimensional data such as grid bus voltage / current, transmission line power, new energy output, energy storage system charging and discharging status, and load power consumption characteristics.

[0009] Data preprocessing unit: Built-in FFT harmonic analysis algorithm module, which performs preprocessing operations such as filtering and denoising, data standardization, missing value completion, and outlier removal on the acquired raw data; accurately decomposes the 2nd to 7th harmonics in the voltage / current signal through the FFT algorithm, and extracts key characteristic parameters such as harmonic phase and amplitude.

[0010] Data storage unit: Adopts a distributed database architecture to support the classified and hierarchical storage of massive amounts of measured data and technical information.

[0011] Electromagnetic Transient Simulation Modeling Module: Constructs an electromagnetic transient simulation model of a power grid-load-storage system that closely matches actual engineering conditions, enabling accurate modeling and dynamic adaptive optimization of core components such as the power grid, new energy sources, energy storage, and loads. It includes a basic topology modeling unit, a new energy and energy storage model integration unit, a load disturbance modeling unit, a dynamic adaptive optimization unit, a multi-element coupled dynamic modeling unit for power grid-load-storage systems, and a model library unit.

[0012] Basic topology modeling unit: Build an electromagnetic transient level system model covering power grid, transmission lines, busbars, and transformers in the simulation platform.

[0013] New energy and energy storage model integration unit: Import black box models provided by wind power, photovoltaic, energy storage, SVG and other equipment manufacturers, and complete the electrical connection configuration between the model and the system basic topology.

[0014] Load disturbance modeling unit: Based on RLC passive components, construct the load equivalent circuit, and write time-varying impedance control logic in C language or simulation script to accurately reproduce the start-up impact, power fluctuation and harmonic emission characteristics of industrial loads; carry out special modeling for typical industrial loads.

[0015] Dynamic adaptive optimization unit: Embedded with LSTM network self-learning algorithm, constructing a feedback mechanism to compare simulation output and actual characteristics, and comparing load model output data with field measured characteristic data in real time.

[0016] The multi-dimensional coupled dynamic modeling unit for power generation, grid, load, and storage constructs a multi-dimensional coupled model encompassing electrical, physical, and environmental dimensions. By introducing digital twin technology, it establishes a real-time mapping between the physical entities of new energy generating units, energy storage batteries, and industrial loads and the simulation model, feeding back measured physical state parameters to the simulation model in real time.

[0017] Model Library Unit: Construct a typical model library covering multiple types of new energy, energy storage, SVG equipment and industrial loads; store standard parameters, equivalent circuit topologies, control logic scripts and measured characteristic curves of various equipment.

[0018] Multi-dimensional simulation accuracy verification module: It comprehensively verifies the simulation model from three core dimensions: electrical characteristics, physical process, and economic performance, and evaluates the model accuracy based on the verification results. It includes electrical dimension verification unit, physical process verification unit, economic dimension verification unit, and verification result evaluation unit.

[0019] Electrical dimension verification unit: Focuses on the core electromagnetic transient indicators of the system and conducts multi-dimensional accuracy verification.

[0020] Physical process verification unit: For core equipment such as transformers and transmission lines, verify the consistency of their loss characteristics with the actual equipment.

[0021] Economic dimension verification unit: Combined with peak-valley-flat electricity pricing mechanism, monitor the economics of energy storage charging and discharging strategies; by comparing the cost per kilowatt-hour in simulation scenarios with the actual cost per kilowatt-hour in engineering projects, evaluate the economic feasibility of control strategies.

[0022] Verification Result Evaluation Unit: Integrates verification data from various dimensions and automatically generates multi-dimensional verification reports.

[0023] The hardware-in-the-loop real-time simulation system establishes a reliable communication link between the simulation computing server and the energy management controller, enabling bidirectional real-time interaction of simulation data and control commands. It includes a communication configuration unit, a virtual communication mapping unit, an anti-interference processing unit, a delay adaptive compensation unit, a precision real-time balance control module, and a data interaction unit.

[0024] Communication configuration unit: Supports local area network configuration between simulation computing server and energy management controller.

[0025] Virtual communication mapping unit: Constructs a precise mapping mechanism between internal variables and protocol messages to realize the format conversion between simulation data and control commands.

[0026] Anti-interference processing unit: Adopts a dual anti-interference protection mechanism to ensure secure and reliable data transmission.

[0027] Delay Adaptive Compensation Unit: Based on the ARIMA algorithm, a communication delay prediction model is constructed to predict the communication delay trend over the next 10 simulation steps; the execution time of control commands is dynamically adjusted according to the prediction results to compensate for the impact of communication delay on the simulation real-time performance.

[0028] Accuracy and real-time balance control module: adopts a dynamic control strategy to ensure real-time simulation of core indicators.

[0029] Data interaction unit: Enables high-speed data interaction between the simulation system and the energy management controller.

[0030] Closed-loop simulation verification module: Constructs a closed-loop linkage system between the simulation model and the energy management controller. By configuring different test scenarios, it verifies the effectiveness and reliability of the source-grid-load-storage coordinated control strategy and evaluates the dynamic and static response characteristics of the system under various operating conditions. It includes a control strategy configuration unit, a test scenario generation unit, a real-time simulation operation unit, and a dynamic monitoring unit.

[0031] Control strategy configuration unit: Supports the installation of various source-grid-load-storage coordinated control logics into the energy management controller, including wind and solar power maximization absorption strategy, energy storage peak-valley arbitrage strategy, wind and solar power output fluctuation smoothing strategy, load impact suppression strategy, etc.; key control parameters can be preset.

[0032] Test Scenario Generation Unit: Constructs a library of multiple test scenarios, covering both normal and extreme operating conditions.

[0033] Real-time simulation operation unit: initiates closed-loop linkage between simulation computing server and energy management controller, constructs a real-time loop link for simulation data acquisition, controller command calculation, simulation model adjustment and data feedback; drives simulation operation according to preset test scenarios, coordinates simulation step size and communication frequency in real time, and realistically simulates the actual operation process of source-grid-load-storage system.

[0034] Dynamic monitoring unit: Real-time acquisition of core data from the simulation system and controller, including system voltage, current, power, frequency, energy storage SOC, and controller adjustment commands.

[0035] Simulation Result Engineering Conversion Module: This module transforms simulation verification results into outcomes that can be directly applied to engineering practice. It includes a parameter optimization unit, a fault risk analysis unit, a report generation unit, and a data export unit.

[0036] Parameter optimization unit: The NSGA-Ⅲ multi-objective optimization algorithm is adopted, with voltage stability, optimal economy and longest energy storage life as the core optimization objectives. It automatically optimizes key parameters such as power allocation coefficient, PID adjustment parameters and energy storage charging and discharging threshold of the energy management controller.

[0037] Fault Risk Analysis Unit: Based on various data collected during the simulation process, big data analysis methods are used to identify potential fault risks in the system, including critical nodes prone to voltage sags, abnormal fluctuations in energy storage SOC, and frequency instability risks caused by sudden drops in renewable energy output.

[0038] Report generation unit: Automatically generates engineering simulation reports.

[0039] Data export unit: Supports standardized export of various results such as simulation curve data, optimization parameter tables, multi-dimensional verification reports, and fault risk analysis data.

[0040] This invention provides a real-time simulation method for electromagnetic transients in source-grid-load-storage coordinated control, comprising the following steps.

[0041] Step 1: The data acquisition and preprocessing module performs full-scene data acquisition and preprocessing.

[0042] Step 2: The electromagnetic transient simulation modeling module constructs and optimizes the electromagnetic transient simulation model.

[0043] Step 3: The multi-dimensional simulation accuracy verification module performs multi-dimensional accuracy verification to form a closed-loop correction.

[0044] Step 4: Construct a reliable communication link between the simulation computing server and the energy management controller in the hardware-in-the-loop real-time simulation system.

[0045] Step 5: The closed-loop simulation verification module realizes the closed-loop linkage between the simulation model and the controller to verify the effectiveness of the control strategy.

[0046] Step 6: The simulation results engineering conversion module converts the simulation results into usable engineering outcomes.

[0047] The technical solution of this application has achieved the following beneficial effects.

[0048] 1. Overcoming the limitation of general load models in simulating transient shocks: The method of constructing physical topology based on basic discrete components (resistors, inductors, capacitors) to represent industrial loads is used. By constructing an equivalent load model based on RLC topology, the starting shock and random fluctuation characteristics of industrial loads such as silicon production and iron smelting are realistically reproduced, solving the problem that traditional models cannot assess the dynamic response capability of controllers.

[0049] 2. Multi-dimensional coupled modeling eliminates fragmentation errors: Through the multi-dimensional coupled model of electrical-physical-environment and the real-time mapping link of digital twin, the physical state of the equipment, the environmental conditions and electrical characteristics are adapted in a coordinated manner, avoiding the simulation deviation caused by single-dimensional modeling, and making the model closer to the actual operating characteristics of the project.

[0050] 3. Time-varying load modeling reproduces real disturbance characteristics: Based on the hybrid modeling architecture of CNN-LSTM fusion model and time-varying impedance control logic, it can accurately reproduce the time-varying characteristics and transient impacts during the start-up and operation of industrial loads, solve the problem that traditional fixed parameter load models cannot simulate real load disturbances, and improve the reliability of control strategy verification under extreme conditions. Attached Figure Description

[0051] Figure 1This is a schematic diagram of the electromagnetic transient real-time simulation system for source-grid-load-storage coordinated control in the embodiments of this application.

[0052] Figure 2 This is a schematic diagram of a hardware-in-the-loop real-time simulation system for the electromagnetic transient real-time simulation system of source-grid-load-storage coordinated control in the embodiments of this application. Detailed Implementation

[0053] The present application will now be further described with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present application and should not be construed as limiting the scope of protection of the present application. It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present application.

[0054] This invention aims to provide a simulation verification method for an integrated source-grid-load-storage system compatible with different load types without the need for a physical interface. By constructing a high-precision electromagnetic transient load model as the main disturbance source, a virtual communication mapping layer is built inside the electromagnetic transient simulation environment based on a virtual communication interface. The system is directly connected to the real controller via an Ethernet port, and the controller's adjustment capability and robustness in the face of complex load shocks and new energy fluctuations are verified in a laboratory environment.

[0055] like Figure 1 and Figure 2 As shown, this invention discloses a source-grid-load-storage coordinated control electromagnetic transient real-time simulation system, comprising:

[0056] The data acquisition and preprocessing module enables accurate acquisition, standardized preprocessing, and secure and reliable storage of data across the entire source-grid-load-storage system. This provides high-quality data support for subsequent simulation modeling, accuracy verification, and strategy validation, ensuring the integrity, accuracy, and timeliness of the input data and resolving the disconnect between simulation input data and actual engineering data. It includes a data acquisition unit, a data preprocessing unit, and a data storage unit.

[0057] Data acquisition unit: Includes high-precision power quality analyzer, distributed field measurement terminal, sensor network and other equipment to realize real-time acquisition of multi-dimensional data such as grid bus voltage / current, transmission line power, new energy (wind power / photovoltaic) output, energy storage system charging and discharging status, load power consumption characteristics, etc. The acquisition frequency can be configured according to simulation requirements to ensure that details such as transient impacts and harmonics are captured.

[0058] Data preprocessing unit: Built-in FFT harmonic analysis algorithm module, which performs preprocessing operations such as filtering and denoising, data standardization, missing value completion, and outlier removal on the collected raw data; accurately decomposes the 2nd to 7th harmonics in voltage / current signals through FFT algorithm, extracts key characteristic parameters such as harmonic phase and amplitude, and provides harmonic data support for load modeling and accuracy verification.

[0059] Data storage unit: Adopting a distributed database architecture, it supports the classified and hierarchical storage of massive amounts of measured data and technical information (equipment parameter manuals, manufacturer black box model descriptions, project main wiring diagrams, etc.); It has data indexing and fast retrieval functions, and can quickly retrieve data according to data type, collection time, equipment number and other dimensions. At the same time, it supports data backup and disaster recovery to ensure data security.

[0060] Data preprocessing adopts a hierarchical processing rule: the raw measured data is processed by the data acquisition preprocessing module, and the simulation data is processed by the fault risk analysis unit. The outlier removal thresholds for both are unified, such as the 3σ criterion, but σ is 0.05 for measured data and 0.1 for simulation data. Parameter optimization adopts a time-series collaborative rule: the model parameters are first optimized by the dynamic adaptive optimization unit, and then the optimized model parameters are used as the initial values ​​for the parameter optimization unit to carry out controller parameter optimization, thus avoiding parameter conflicts.

[0061] Electromagnetic Transient Simulation Modeling Module: This module constructs an electromagnetic transient simulation model of a power grid-load-storage system that closely matches actual engineering conditions. It enables accurate modeling and dynamic adaptive optimization of core components such as the power grid, new energy sources, energy storage, and loads, ensuring the model accurately reproduces the electromagnetic transient characteristics under various operating conditions, including normal system operation, load disturbances, and new energy fluctuations. The module includes a basic topology modeling unit, a new energy and energy storage model integration unit, a load disturbance modeling unit, a dynamic adaptive optimization unit, a multi-element coupling dynamic modeling unit for power grid-load-storage systems, and a model library unit.

[0062] Basic Topology Modeling Unit: Based on the target project's main wiring topology drawings and equipment parameters, an electromagnetic transient-level system model covering the power grid, transmission lines, busbars, and transformers is built in the simulation platform. During modeling, the actual connection relationships and parameter configurations of the equipment are strictly followed. Topology consistency verification ensures that the model's topology structure completely matches the actual project, avoiding simulation distortion due to topology deviations. Specifically, the modeling process first requires obtaining basic data such as the target project's main wiring topology structure, component equipment parameter information, and black-box models provided by the equipment manufacturers. Then, based on the obtained actual parameters and data, the electromagnetic transient-level system topology is built in the ADPSS software.

[0063] The new energy and energy storage model integration unit imports black-box models provided by wind power, photovoltaic, energy storage, and SVG equipment manufacturers, and completes the electrical connection configuration between the model and the system's basic topology. It optimizes model characteristics using offline simulation and parameter tuning, conducting offline electromagnetic transient simulation calculations on a local computer and monitoring key indicators such as model terminal voltage, active power, and reactive power in real time. Based on the simulation results, it dynamically adjusts parameters such as the SVG reactive power setpoint and the per-unit value of the infinite power supply line voltage, iterating repeatedly until the model's output characteristics match the actual operating characteristics of the equipment, ensuring that the model accurately reflects core characteristics such as new energy output fluctuations and energy storage charging and discharging responses. Offline simulation is a method of electromagnetic transient simulation calculation based on a local computer. The specific steps for parameter tuning are as follows: after starting the simulation, wait for the simulation results to be generated, and observe whether the active power, reactive power, and terminal voltage of the wind power, photovoltaic, energy storage, and SVG unit models meet the normal standards, especially whether the per-unit value of the terminal voltage is between 0.95pu and 1.05pu. Based on the simulation results, adjust the SVG reactive power setting value and the per-unit value of the infinite power line voltage until the terminal voltage of the wind, solar, and energy storage SVG models are all within the normal range and the active power ramp-up response is normal.

[0064] Load disturbance modeling unit: Based on RLC passive components, construct the load equivalent circuit, and write time-varying impedance control logic in C language or simulation script to accurately reproduce the start-up impact, power fluctuation and harmonic emission characteristics of industrial loads (such as electric arc furnaces, rolling mills, etc.); carry out special modeling for typical industrial loads.

[0065] Dynamic Adaptive Optimization Unit: Embedded with an LSTM network self-learning algorithm, it constructs a feedback mechanism for comparing simulation output and actual characteristics, comparing load model output data with field measured characteristic data in real time. When the deviation exceeds 5%, the algorithm automatically adjusts the RLC parameter time series function coefficients to ensure a model matching degree ≥95%. Simultaneously, it calls upon typical parameter curves of various types of industrial loads (such as electric arc furnaces, water pumps, and fans) from the model library to achieve rapid and accurate modeling of similar loads, improving modeling efficiency. It also assists in completing the simulation verification and parameter tuning of voltage, current, and power at key system nodes, ensuring that the overall characteristics of the model are consistent with the actual engineering.

[0066] The source-grid-load-storage multi-dimensional coupled dynamic modeling unit overcomes the limitations of existing independent component modeling and static parameter matching, constructing a multi-dimensional coupled model of electrical, physical, and environmental dimensions. By introducing digital twin technology, a real-time mapping is established between the physical entities of new energy units, energy storage batteries, and industrial loads and the simulation model. Measured physical state parameters are fed back to the simulation model in real time, dynamically correcting the model parameters. Based on measured wind speed, illumination, and ambient temperature data, a coupled influence model is constructed to achieve accurate simulation under extreme environmental conditions.

[0067] Model Library Unit: This unit constructs a typical model library covering various types of new energy sources, energy storage (lithium batteries, vanadium redox flow batteries, etc.), SVG equipment, and industrial loads (electric arc furnaces, rolling mills, injection molding machines, etc.). It stores standard parameters, equivalent circuit topologies, control logic scripts, and measured characteristic curves for various equipment, supporting rapid model recall, modification, and expansion. It also records historical data on model parameter tuning. The various models and parameters stored in the model library can significantly improve modeling efficiency and accuracy.

[0068] The multi-dimensional simulation accuracy verification module comprehensively verifies the simulation model from three core dimensions: electrical characteristics, physical processes, and economic performance. It evaluates the model's accuracy based on the verification results, forming a closed-loop optimization mechanism of verification, feedback, and correction. This ensures that the simulation model accurately reflects the actual operating state of the power grid-load-storage system, providing reliable model support for control strategy verification. It includes an electrical dimension verification unit, a physical process verification unit, an economic dimension verification unit, and a verification result evaluation unit.

[0069] Electrical Dimension Verification Unit: Focusing on the core electromagnetic transient indicators of the system, multi-dimensional accuracy verification is carried out: 1) Voltage sag / surge verification, ensuring that the error between the simulated value and the measured value is ≤3%; 2) Harmonic characteristic verification, focusing on verifying the phase and amplitude of the 2nd to 7th harmonics to ensure that the harmonic characteristics are consistent with the actual situation; 3) Inrush current verification, verifying the rise time of the inrush current, with an error of ≤2ms; Through the above verification, it is ensured that the electrical characteristics of the model are highly consistent with the actual system on site.

[0070] Physical process verification unit: Based on the principle of energy conservation, physical characteristics are verified. For core equipment such as transformers and transmission lines, the consistency of their loss characteristics (such as copper loss and iron loss) with the actual equipment is verified. For energy storage systems, the focus is on verifying the charging and discharging efficiency and the accuracy of SOC state of charge estimation to ensure that the model can truly reflect the energy conversion process of the energy storage system.

[0071] Economic dimension verification unit: Combined with peak-valley-flat electricity pricing mechanism, monitor the economics of energy storage charging and discharging strategies; evaluate the economic feasibility of control strategies by comparing the cost per kilowatt-hour in simulation scenarios with the actual cost per kilowatt-hour in engineering projects; provide data support for verifying economically optimal scheduling decisions, and ensure that simulation results can guide the economic operation optimization of actual projects.

[0072] Verification Result Evaluation Unit: Integrates verification data from various dimensions and automatically generates a multi-dimensional verification report. The report includes key information such as verification indicators, simulation values, measured values, error range, and whether the standards are met. If any indicator fails to meet the standards, it is automatically fed back to the electromagnetic transient simulation modeling module, triggering the model parameter correction process until all verification indicators meet the requirements and the model accuracy meets the standards before proceeding to the subsequent simulation verification stage.

[0073] The hardware-in-the-loop (HIL) real-time simulation system establishes a reliable communication link between the simulation computing server and the energy management controller, wind farm energy management controller, photovoltaic energy management controller, and energy storage energy management controller. This enables bidirectional real-time interaction of simulation data and control commands, resolving communication compatibility issues between the simulation system and actual control equipment, and providing communication assurance for HIL simulation. It also possesses anti-interference and delay compensation capabilities to ensure communication stability and data transmission accuracy. Specific hardware includes a simulation computing server, energy management controller, wind farm energy management controller, photovoltaic energy management controller, energy storage energy management controller, monitor, and network switch. The network switch connects the simulation computing server and the energy management controller to form a local area network, achieving reliable networking at the hardware level. A virtual communication interface module is loaded into the simulation computing server to establish a mapping mechanism between internal variables and protocol messages. The system includes a communication configuration unit, a virtual communication mapping unit, an anti-interference processing unit, a delay adaptive compensation unit, a precision real-time balance control module, and a data interaction unit.

[0074] Communication configuration unit: Supports local area network configuration of simulation calculation server with energy management controller, wind farm energy management controller, photovoltaic energy management controller and energy storage energy management controller. The server IP address (such as 192.168.1.10), controller IP address (such as 192.168.1.20) and communication port number can be flexibly set; built-in network connectivity detection tool can automatically detect link connection status. If communication abnormality is detected, an alarm will be set in time to ensure network reliability.

[0075] Virtual communication mapping unit: Constructs a precise mapping mechanism between internal variables and protocol messages to realize the format conversion between simulation data and control commands.

[0076] Forward transmission: Real-time reading of memory variable values ​​such as PCC common connection point voltage, current, energy storage SOC, and new energy output within the simulation step size, and calling the IEC 104 / Modbus TCP / IEC 61850 protocol stack to package them into binary messages adapted to energy management controllers, wind farm energy management controllers, photovoltaic energy management controllers, and energy storage energy management controllers.

[0077] Reverse transmission: Receive control command messages from the energy management controller, wind farm energy management controller, photovoltaic energy management controller, and energy storage energy management controller, unpack them into floating-point format within the module, and transmit them to the control input element of the simulation model to complete the precise driving of the control commands.

[0078] Anti-interference processing unit: Employs a dual anti-interference protection mechanism to ensure secure and reliable data transmission. It uses AES-128 encryption to encrypt transmitted data and employs CRC-32 redundancy check to verify data integrity, preventing data tampering and loss. It is configured with primary and backup dual-network redundant links, monitoring primary network communication latency in real time. If the primary network latency exceeds 10ms or a failure occurs, it automatically switches to the backup network within 50ms. After switching, it marks the faulty link type and records fault information in real time, ensuring uninterrupted communication continuity.

[0079] Delay Adaptive Compensation Unit: Based on the ARIMA algorithm, a communication delay prediction model is constructed to predict the communication delay trend over the next 10 simulation steps. The execution time of control commands is dynamically adjusted according to the prediction results to compensate for the impact of communication delay on the simulation real-time performance. A 50ms delay threshold is set. When the actual communication delay exceeds the threshold, an emergency control logic simulation verification is automatically triggered to evaluate the stability of the system under delay conditions. At the same time, a delay alarm signal is output and the simulation step size is temporarily adjusted to ensure the simulation real-time performance.

[0080] Accuracy and real-time balance control module: adopts a dynamic control strategy to ensure real-time simulation of core indicators.

[0081] Data Interaction Unit: Enables high-speed data interaction between the simulation system and energy management controllers, wind farm energy management controllers, photovoltaic energy management controllers, and energy storage energy management controllers. It supports configuring data transmission frequency according to simulation requirements; features data caching to temporarily store data during transmission, preventing data loss due to instantaneous network fluctuations; provides real-time feedback of CRC-32 verification results after data transmission, automatically retransmitting if verification fails; and records data interaction logs including data transmission time, data content, transmission status, transmission volume, and error values, facilitating subsequent fault diagnosis and link optimization. The display shows real-time network status, communication latency, data interaction logs, and abnormal alarm information, enabling visualized monitoring of the communication link.

[0082] Closed-loop simulation verification module: This module constructs a closed-loop linkage system between the simulation model and the energy management controller, wind farm energy management controller, photovoltaic energy management controller, and energy storage energy management controller. By configuring different test scenarios, it verifies the effectiveness and reliability of the source-grid-load-storage coordinated control strategy, evaluates the dynamic and static response characteristics of the system under various operating conditions, and provides experimental basis for control strategy optimization and parameter tuning. It includes a control strategy configuration unit, a test scenario generation unit, a real-time simulation operation unit, and a dynamic monitoring unit.

[0083] Control strategy configuration unit: Supports the installation of various source-grid-load-storage coordinated control logics onto the energy management controller, including wind and solar power maximization absorption strategy, energy storage peak-valley arbitrage strategy, wind and solar power output fluctuation smoothing strategy, load impact suppression strategy, etc.; supports the installation of wind power output regulation strategy onto the wind farm energy management controller, photovoltaic maximum power point tracking and output smoothing control strategies onto the photovoltaic energy management controller, and energy storage charge and discharge control and SOC balancing control strategies onto the energy storage energy management controller; allows preset key control parameters for each controller, such as energy storage SOC charge and discharge threshold (30%-95%), power regulation dead zone, response delay, etc., and supports online modification and optimization of control parameters.

[0084] Test Scenario Generation Unit: Constructs a multi-type test scenario library covering both conventional and extreme operating conditions: 1) Basic test scenarios: including renewable energy output fluctuation scenarios, load step response scenarios, and electricity price economic dispatch scenarios; 2) Complex test scenarios: including cross-scenario collaborative scenarios (connecting to the upstream power grid and surrounding user models to simulate regional-level source-grid-load-storage interaction) and extreme operating condition scenarios (including six types of operating conditions such as sudden drop in renewable energy output, simultaneous impact of multiple loads, communication interruption, and equipment failure); preset assessment indicators for each test scenario, such as voltage recovery time ≤ 2s, frequency fluctuation ≤ ±0.2Hz, and energy storage charging and discharging efficiency ≥ 90%.

[0085] Real-time simulation operation unit: Starts closed-loop linkage between simulation calculation server and energy management controller, wind farm energy management controller, photovoltaic energy management controller, and energy storage energy management controller, and constructs a real-time loop link for simulation data acquisition, controller command calculation, simulation model adjustment and data feedback; drives simulation operation according to preset test scenarios, coordinates simulation step size and communication frequency in real time to ensure the real-time performance and stability of the simulation process, and realistically simulates the actual operation process of the source-grid-load-storage system.

[0086] Dynamic monitoring unit: Real-time acquisition of core data from the simulation system and controller, including system voltage, current, power, frequency, energy storage SOC, and controller adjustment commands; intuitive display of data change trends through dynamic curves and data tables, and real-time presentation of the execution effect of control strategies; equipped with an anomaly alarm function, when the monitored indicators exceed preset thresholds, such as excessive voltage sag or excessive frequency fluctuation, the abnormal event is automatically marked and the relevant data is recorded for subsequent analysis.

[0087] The simulation results engineering transformation module transforms simulation verification results into outcomes directly applicable to engineering practice. Through functions such as parameter optimization, risk analysis, and report generation, it provides precise guidance for the design review, on-site implementation, and operational optimization of power generation, grid, load, and storage projects, effectively implementing simulation results into engineering applications. It includes a parameter optimization unit, a fault risk analysis unit, a report generation unit, and a data export unit.

[0088] Parameter optimization unit: Employing the NSGA-Ⅲ multi-objective optimization algorithm, with voltage stability, optimal economy, and longest energy storage life as core optimization objectives, it automatically optimizes key parameters such as the power allocation coefficient and PID control parameters of the energy management controller, the output limiting coefficient of the wind farm energy management controller, the MPPT control parameters and output smoothing coefficient of the photovoltaic energy management controller, and the charging and discharging threshold and equalization control parameters of the energy storage energy management controller. After optimization, it generates configuration files that can be directly imported into each controller, avoiding the subjectivity and error of manual parameter tuning and improving the engineering adaptability of the control strategy.

[0089] Fault Risk Analysis Unit: Based on various data collected during the simulation process, big data analysis methods are used to identify potential fault risks in the system, including key nodes prone to voltage sags, abnormal fluctuations in energy storage SOC, and frequency instability risks caused by sudden drops in renewable energy output. For the identified risk points, early warning thresholds and targeted prevention and control measures are output, such as adding reactive power compensation devices and optimizing energy storage charging and discharging strategies, to ensure the safe operation of the system.

[0090] Report generation unit: Automatically generates engineering simulation reports, covering core content such as simulation condition descriptions, control strategy effectiveness evaluation, parameter tuning results, fault risk warnings, and achievement status of assessment indicators; using a combination of charts, such as simulation curve comparison charts, parameter optimization trend charts, and risk distribution heat maps, to intuitively present the results, which can be directly used for project design reporting and on-site implementation guidance.

[0091] Data export unit: Supports standardized export of various results such as simulation curve data, optimization parameter tables, multi-dimensional verification reports, and fault risk analysis data; export formats include common formats such as Excel, PDF, and CSV, which facilitates subsequent data archiving, secondary analysis, and engineering document preparation.

[0092] Furthermore, the accuracy and real-time performance balancing control module dynamically balances simulation accuracy and real-time performance through a closed-loop logic of latency monitoring, strategy decision-making, parameter adjustment, and effect feedback, ensuring uninterrupted simulation and consistent monitoring of core indicators even under extreme communication conditions. Specifically, it includes a latency monitoring subunit, a strategy decision-making subunit, a parameter execution subunit, and an effect feedback subunit, all working together to achieve the balancing control function. The latency monitoring subunit reuses existing communication latency data from the hardware-in-the-loop real-time simulation system, reducing system integration costs.

[0093] The delay monitoring subunit collects the actual delay data of the communication link every simulation step (10μs-20μs) and records the delay fluctuation trend (delay standard deviation of 10 consecutive steps) simultaneously.

[0094] The strategy decision-making subunit matches the corresponding accuracy-real-time control strategy based on the preset delay classification threshold. The specific classification and strategy are as follows: a) Optimal accuracy mode (communication delay ≤ 30ms): adopts a 100kHz high-frequency data acquisition frequency and a 10μs simulation step size to fully retain the details of transient impacts, harmonics, etc., and ensure simulation accuracy; b) Balanced mode (30ms < communication delay ≤ 50ms): automatically adjusts to a 50kHz acquisition frequency and a 20μs simulation step size to improve the simulation real-time performance while retaining core transient characteristics (such as voltage sags and inrush currents); c) Emergency protection mode (communication delay > 50ms): triggers emergency control logic to suspend the acquisition and simulation of sub-low frequency harmonics and non-critical node transient details, and only retains the monitoring and simulation of core indicators such as system voltage, frequency, energy storage SOC, and total active power. The simulation step size can be flexibly adjusted to 50μs-100μs. At the same time, the core indicator data caching function is activated to avoid data loss. The strategy decision-making subunit adopts a rule-based decision-making algorithm with a decision delay of ≤1 simulation step (≤20μs) to ensure timely control.

[0095] The parameter execution subunit is linked with the simulation computing server and data acquisition unit through the API interface, and the adjustment command response time is ≤50μs.

[0096] The effect feedback subunit compares the communication delay and simulation accuracy after adjustment in real time. If the delay is still greater than 50ms after adjustment, a communication link failure alarm is automatically triggered and pushed to the anti-interference processing unit of the hardware-in-the-loop real-time simulation system. The main and backup dual-network switching function is activated in coordination to further ensure communication stability.

[0097] This invention provides a real-time simulation method for electromagnetic transients in source-grid-load-storage coordinated control, comprising the following steps.

[0098] Step 1: The data acquisition and preprocessing module performs full-scene data acquisition and preprocessing.

[0099] Step 2: The electromagnetic transient simulation modeling module constructs and optimizes the electromagnetic transient simulation model.

[0100] Step 3: The multi-dimensional simulation accuracy verification module performs multi-dimensional accuracy verification to form a closed-loop correction.

[0101] Step 4: Construct a reliable communication link between the simulation computing server and the energy management controller in the hardware-in-the-loop real-time simulation system.

[0102] Step 5: The closed-loop simulation verification module realizes the closed-loop linkage between the simulation model and the controller to verify the effectiveness of the control strategy.

[0103] Step 6: The simulation results engineering conversion module converts the simulation results into usable engineering outcomes.

[0104] Example 1.

[0105] This embodiment 1 discloses a real-time simulation method for electromagnetic transients of source-grid-load-storage coordinated control, wherein step 1 includes the following steps.

[0106] 1.1 The data acquisition unit starts the integrated high-precision power quality analyzer, distributed field measurement terminals, and sensor network. It adopts a spatiotemporal collaborative acquisition mode. In addition to real-time acquisition of basic data such as grid bus voltage / current, transmission line power, wind power / photovoltaic output, energy storage charging and discharging status, and industrial load power consumption characteristics, it additionally acquires related physical quantities such as the pitch angle / photovoltaic panel tilt angle of new energy units (wind power / photovoltaic), the voltage / temperature of individual battery cells in the energy storage system, and furnace condition parameters (such as furnace temperature and electrode position) of industrial loads (such as electric arc furnaces). The acquisition frequency is dynamically configured according to the simulation's need for transient details, and the environmental parameters at the time of data acquisition are recorded simultaneously to ensure that characteristic data such as load transient impacts and harmonics, as well as the laws of multi-factor coupling influence, are captured.

[0107] 1.2 The data preprocessing unit employs a Fast Fourier Transform (FFT) and Variational Mode Decomposition (VMD) fusion algorithm module. First, the VMD algorithm decomposes the acquired raw voltage / current signal into multiple intrinsic mode components, achieving accurate separation of transient impact signals and steady-state signals. Then, each component is filtered and denoised separately. For non-stationary data caused by new energy fluctuations and load impacts, an attention-based missing value completion algorithm is used, combined with the 3σ criterion and the isolated forest algorithm to remove outliers. The FFT algorithm decomposes the 2nd to 7th harmonics in the voltage / current signal, extracting key parameters such as harmonic phase and amplitude. Simultaneously, the VMD component reconstruction obtains the characteristic parameters of the sub-low frequency harmonics and interharmonics, forming a full-band harmonic dataset.

[0108] 1.3 The data storage unit adopts a hybrid storage architecture of time-series database and distributed file system. The preprocessed measured data is classified and stored according to basic electrical data, related physical quantity data, environmental parameter data, and full-band harmonic data. Among them, high-frequency transient data (≥100kHz) is stored in time-series database to ensure retrieval efficiency, while steady-state operation data and project technical documents (equipment parameter manual, black-box model description, main wiring diagram) are stored in distributed file system. An association index of data, operating conditions, and model parameters is established to enable rapid retrieval of corresponding data according to simulation scenarios (such as sudden drop in new energy and load impact). It has a data quality classification and labeling function, which labels the data into three levels: A, B, and C (level A is core simulation data, level B is auxiliary verification data, and level C is reference data) according to data integrity, accuracy, and acquisition scenario. At the same time, the core measured data (such as transient impact waveform) is hashed and encrypted using blockchain technology to ensure that the data is tamper-proof and to provide traceable data quality assurance for subsequent accuracy verification of simulation models.

[0109] Example 2.

[0110] This embodiment 2 discloses a real-time simulation method for electromagnetic transients of source-grid-load-storage coordinated control, wherein step 2 includes the following steps.

[0111] 2.1 The model library unit constructs an intelligent model library with intelligent retrieval, dynamic updates, and knowledge transfer. Based on semantic retrieval, it can quickly match typical models of similar projects by inputting keywords such as project type, load characteristics, and new energy installed capacity. A dynamic update mechanism for model parameters is established, which associates and stores the parameter tuning data, operating condition adaptation results, and measured data of this modeling with the measured data. The modeling data of multiple projects is integrated through federated learning algorithms to optimize the universality of typical parameters in the model library. For similar projects, the validated model parameters in the model library can be migrated to the new model, requiring only a small number of iterative corrections based on the measured data of the new project, which greatly improves the modeling efficiency.

[0112] 2.2 The basic topology modeling unit adopts an intelligent modeling mode of intelligent drawing parsing and parameter adaptive matching. Based on the project's main wiring diagram (PDF or CAD format), it automatically extracts topology nodes, equipment types, and connection relationships through deep learning image recognition algorithms to generate a structured topology description file. Combined with the collected actual measured parameters of the equipment, the default equipment parameters in the ADPSS software are corrected to build an electromagnetic transient level system topology covering the power grid, transmission lines, busbars, and transformers. A topology and parameter linkage verification mechanism is adopted. Multiple sets of parameter disturbance scenarios are generated through Monte Carlo simulation to verify the stability of the topology under parameter fluctuations, ensuring complete matching with the actual engineering connection relationships and operating characteristics.

[0113] 2.3 The source-grid-load-storage multi-dimensional coupling dynamic modeling unit constructs a multi-dimensional coupling model of electrical, physical and environmental dimensions.

[0114] 2.4. The new energy and energy storage model integration unit addresses the poor compatibility of manufacturers' black-box models by employing an adaptive conversion module for black-box model interfaces. This module automatically identifies the interface protocols (such as DLL and FMU) of different manufacturers' black-box models and converts them to an ADPSS-compatible interface format. After completing the electrical connection between the model and the basic topology, a two-stage parameter tuning mode of offline simulation and online prediction is initiated: The first stage, based on collected new energy output fluctuation data and energy storage charging and discharging time-series data, uses a clustering algorithm to divide typical operating conditions (such as wind power gusts, photovoltaic shading, and energy storage peak-valley charging and discharging) and conducts offline simulations for each operating condition on the local PC. The second stage embeds the Gradient Boosting Tree (GBT) algorithm, predicts the model output deviation for different parameter combinations based on offline simulation data, and specifically adjusts parameters such as the SVG reactive power setpoint, energy storage charging and discharging response coefficient, and the per-unit value of the infinite power line voltage. Iterative optimization continues until the model's terminal voltage (0.95 pu-1.05 pu) and active / reactive power response deviations under all typical operating conditions are ≤2%, achieving consistency between the model and the actual equipment's full-condition characteristics.

[0115] The black-box model interface adaptive conversion module solves the interface incompatibility problem between black-box models from different manufacturers and the ADPSS simulation platform, enabling plug-and-play use of the black-box model while ensuring the electrical connection consistency between the model and the basic topology and the real-time interaction of simulation data. It includes an interface protocol identification subunit, a protocol conversion subunit, an adaptation verification subunit, and an anomaly handling subunit.

[0116] Interface Protocol Identification Subunit: Automatically scans and identifies the interface type and protocol specification of the black-box model, supports mainstream formats (DLL dynamic link library, FMU functional model unit, S-Function, etc.), and extracts interface parameters (input / output variable type, data precision, call timing requirements).

[0117] Protocol conversion subunit: Based on the identified black-box model interface protocol, it converts it into an interface format compatible with the ADPSS simulation platform (such as the custom model interface supported by ADPSS, CSV data interaction interface), realizing the adaptation of data format (binary / floating-point / string) and communication timing.

[0118] Adaptation and verification subunit: After completing the protocol conversion, it automatically performs interface connectivity testing, data transmission accuracy verification (such as the error between the voltage and power data output by the model and the simulation platform ≤ 2%), and real-time testing (interface call delay ≤ 10μs) to ensure that the converted model can be normally connected to the simulation system.

[0119] The exception handling subunit outputs clear alarm information for issues such as interface identification failure, data transmission packet loss, and protocol incompatibility, such as an unidentified FMU version or DLL function call failure, and provides adaptation suggestions, such as upgrading the FMU parsing library or supplementing the black-box model interface documentation.

[0120] The workflow is as follows: Users import black-box models (including DLL files, FMU packages, and interface documentation) provided by new energy and energy storage manufacturers into the module; the interface protocol identification subunit identifies the interface type and extracts core parameters by parsing the model file header information and calling preset protocol parsing rules; the protocol conversion subunit generates an adapted conversion intermediate layer based on the ADPSS interface specification: for DLL type models, it encapsulates function interfaces adapted to ADPSS calls to achieve simulation step size synchronization and input / output variable mapping; for FMU type models, it parses the FMU's modelDescription.xml file, extracts model variable metadata, converts it to variable naming rules supported by ADPSS, and adapts the FMU's Co-Simulation mode to the ADPSS real-time simulation timing; the adaptation verification subunit starts offline debugging, simulates the ADPSS simulation environment to call the converted black-box model, and verifies data transmission accuracy and interface stability. The verified black-box model is then electrically connected to the basic topology model (power grid, transmission lines, etc.) through the converted interface, and the model association relationships in the topology description file are updated synchronously.

[0121] 2.5 The load disturbance modeling unit breaks through the limitations of the traditional single RLC equivalent circuit, constructing a hybrid modeling architecture of RLC passive components and data-driven equivalent models. For industrial loads (such as electric arc furnaces), based on the coupled physical and electrical data such as furnace temperature and electrode position collected in step 1, the nonlinear dynamic characteristics of the load are extracted through a CNN-LSTM fusion model to generate characteristic compensation coefficients. Then, a basic equivalent circuit is constructed based on RLC passive components, and time-varying impedance control logic is written in C language or simulation scripts. The voltage balance equation of the time-varying RLC circuit is established according to Kirchhoff's Voltage Law (KVL), and the voltage balance equation model is as follows.

[0122] .

[0123] In the formula, u(t) is the real-time voltage at the load end; i(t) is the real-time current at the load end; R(t) is the time-varying resistance (single); L(t) is the time-varying inductance; the formula accurately describes the coupling relationship between the time-varying impedance characteristics and the electrical response.

[0124] The time-varying impedance control model is as follows.

[0125] .

[0126] .

[0127] In the formula, R start R is the initial resistance at the load start-up moment; a is the resistance attenuation coefficient; t is the real-time moment; t0 is the load start-up moment; R stable The resistance is the resistance when the load is running stably; ΔR is the feature compensation coefficient output by the CNN-LSTM model, in ohms; L start β is the initial inductance at the moment of load start-up; β is the inductance temperature influence coefficient, dimensionless, with a value range of 0-1; T(t) is the real-time furnace temperature under load; T0 is the initial furnace temperature under load; T max Rated furnace temperature under load; L stable Inductance for stable load operation.

[0128] A remaining life model for the energy storage battery is constructed to verify the consistency between the energy storage life degradation characteristics in the simulation model and those of the actual battery. The remaining life model for the energy storage battery is as follows.

[0129] In the formula, L represents the remaining lifespan of the energy storage battery, in years, i.e., after N years. cyc The battery still has an effective lifespan after one charge-discharge cycle; L base The reference lifespan of the energy storage battery refers to the rated lifespan specified by the battery manufacturer under standard operating conditions (25℃, 80% depth of charge / discharge); e is a natural constant, dimensionless; m is the lifespan degradation coefficient, characterizing the degree of lifespan degradation per unit charge / discharge cycle, obtained by fitting measured battery cycle data; N cyc The cumulative charge-discharge cycle count for the energy storage battery is obtained by statistical analysis of simulated charge-discharge time-series data using the rainflow counting method (DoD ≥ 10% is considered one valid cycle); D DoD ξ3 is the depth-of-charge / discharge influence factor, dimensionless, quantifying the amplification effect of depth-of-charge / discharge on lifetime decay; ξ3 is the temperature influence correction coefficient, dimensionless, ranging from 0.005 to 0.01, characterizing the degree of correction for lifetime decay per unit temperature deviation; ΔT bat This is the difference between the actual temperature and the standard temperature of the energy storage battery.

[0130] The load transient impact and full-band harmonic data extracted in step 1 are substituted into the logic, and the drive circuit generates the same closing surge current (1.5-3 times the rated value), reactive power impact waveform, and 2nd-7th harmonic and interharmonic output as on site, so as to realize the dual reproduction of the electrical and physical characteristics of the load.

[0131] 2.6 The dynamic adaptive optimization unit constructs a collaborative algorithm system of LSTM self-learning and multi-objective constraint optimization. Besides real-time comparison of load model output data with actual field characteristics, it imposes voltage stability constraints (node ​​voltage fluctuation ≤ ±5%), frequency safety constraints (system frequency fluctuation ≤ ±0.2Hz), and energy storage lifetime constraints (charge-discharge cycle count matches the actual lifetime curve). When the deviation between the model output and the measured characteristics exceeds 5%, or any constraint condition is violated, the improved particle swarm optimization (IPSO) algorithm is used to simultaneously optimize the time-series function coefficients and feature compensation coefficient weights of the RLC parameters. This ensures a model matching degree ≥ 95% while also considering the overall stability of the simulation system and the consistency of component characteristics. The improved IPSO model is as follows.

[0132] , .

[0133] In the formula, F is the overall fitness value, dimensionless, and is the core optimization objective of the IPSO algorithm; w1 is the weight coefficient of the model deviation term, dimensionless, with a value range of 0-1. Considering engineering priority settings, a value of 0.6 is preferred to ensure that model matching is the core optimization objective; δ is the relative deviation between the model output and the measured characteristics, dimensionless, with a value range of 0-1; w2 is the weight coefficient of the constraint penalty term, dimensionless, with a value range of 0-1; c is the constraint category index, with values ​​of 1, 2, and 3, corresponding to voltage stability constraints, frequency safety constraints, and energy storage lifetime constraints, respectively; γ c is the penalty coefficient for the c-th constraint, dimensionless, set according to the importance of the constraint, with voltage / frequency constraints being the core safety constraints; viol(c) is the degree of violation of the c-th constraint, dimensionless, 0 when the constraint is satisfied, and the ratio of the actual violation amount to the constraint threshold when the constraint is violated; the optimization objective is to minimize F, ensuring that the model matching degree is ≥95% while taking into account the overall stability of the simulation system and the consistency of component characteristics.

[0134] 2.7 Static Model Verification: Conduct no-load characteristic verification to confirm the transformer no-load current (error ≤ 3%) and voltage ratio (error ≤ 1%); conduct short-circuit characteristic verification to confirm the line short-circuit impedance (error ≤ 2%) and fault current peak value (error ≤ 3%); if the verification fails, return to step 2.1 to remodel; if it passes, proceed to step 3.

[0135] Example 3.

[0136] This embodiment 3 discloses a real-time simulation method for electromagnetic transients of source-grid-load-storage coordinated control, wherein step 3 includes the following steps.

[0137] 3.1 The electrical dimension verification unit constructs a verification system for transient, steady-state, and harmonic full-time-domain electrical characteristics; in addition to verifying the voltage sag / surge amplitude and duration (error ≤3%), the phase and amplitude of the 2nd to 7th harmonics, and the rise time of the impulse current (error ≤2ms), it also verifies the transient voltage oscillation attenuation characteristics through an improved damping coefficient evaluation model.

[0138] .

[0139] ; In the formula, ξ is the improved damping coefficient, dimensionless, characterizing the attenuation characteristics of transient voltage oscillations in the source-grid-load-storage system, and serving as a verification index; ln is the natural logarithm, dimensionless; A k The value represents the k-th adjacent peak value of the transient voltage oscillation curve, in kV, taken from simulated or measured transient voltage data; A k+1 This is the (k+1)th adjacent oscillation peak of the transient voltage oscillation curve, and A k consecutive adjacent and satisfying A k+1 k This conforms to the characteristics of damped oscillation; π is the constant of pi, dimensionless; k1 is the weight of the impact of new energy output fluctuations, dimensionless, with a value range of 0.05-0.1, dynamically adjusted according to the proportion of new energy installed capacity, the higher the proportion, the larger the value, such as 0.1 when the installed capacity proportion is ≥30%, and 0.05 when the proportion is <30%; ΔP renew P is the dimensionless coefficient of the power output fluctuation of the new energy source during the oscillation. renew,k For the actual output of new energy sources (wind power or photovoltaic) during the oscillation period, P renew,0 The rated output of new energy sources; k2 is the load impact intensity correction weight, dimensionless, with a value range of 0.03-0.08. The higher the proportion of industrial load, the larger the value. For example, when the proportion of industrial load is ≥50%, it is 0.08, and when the proportion is <50%, it is 0.03. * imp I represents the per-unit value of the load impulse current, which is dimensionless and characterizes the load impulse intensity. imp I represents the peak value of the actual load inrush current. rated For load rated current and voltage imbalance verification (negative sequence voltage imbalance ≤2%), a dynamic comparison algorithm of measured and simulated curves is adopted. The similarity between the two curves is calculated by dynamic time warping (DTW), avoiding the limitations of traditional point-to-point comparison and accurately capturing subtle differences in transient processes.

[0140] 3.2 The physical process verification unit is based on the principle of energy conservation and combines the coupling characteristics of multiple components such as source, grid, load and storage to construct an improved energy balance verification model for the whole system.

[0141] ​, .

[0142] In the formula, ΔE represents the energy imbalance of the improved system, in MJ, and is a core verification index for measuring the consistency between the simulation model and the energy conservation characteristics of the actual system. The smaller the value, the closer the energy characteristics of the model are to reality; E in The total energy input to the system, encompassing all energy entering the source-grid-load-storage system, including the output energy of new energy sources (wind power or photovoltaic) and the energy input from the grid side, is calculated by integrating the real-time power time-series data of each input unit collected in step 1. E out The total output energy of the system, encompassing energy consumed by industrial and residential loads, as well as all energy leaving the system, including energy fed back to the grid, is calculated by integrating real-time output power time-series data; E loss The total energy loss of conventional components such as lines and transformers, including transmission line resistance heating losses, transformer iron and copper losses, and switchgear contact losses, is calculated based on the equipment parameters (such as line impedance and transformer loss coefficient) modeled in step 2; ΔE bat ξ2 is the energy correction term for additional losses during energy storage charging and discharging, in MJ, characterizing the additional losses caused by voltage fluctuations and polarization effects during the charging and discharging process of the energy storage battery; ξ2 is the energy storage additional loss correction coefficient, dimensionless, ranging from 0.02 to 0.05, with a larger value for the greater the depth of charge and discharge; C is the equivalent capacitance of the energy storage battery, obtained from the measured parameters of the individual energy storage battery cells; U max U min The maximum / minimum single-cell voltage of the energy storage battery during the assessment period is taken from the collected time-series data of the single-cell voltage of the energy storage battery.

[0143] 3.3 The economic dimension verification unit constructs a three-dimensional economic verification system for full life cycle cost (LCC) and multi-scenario benefits; in addition to comparing the simulated cost per kilowatt-hour with the actual project cost (error ≤ 5%) in combination with the peak-valley flat electricity price mechanism, it conducts full life cycle cost decomposition verification and multi-scenario benefit evaluation.

[0144] 3.4 The verification result evaluation unit constructs a closed-loop correction system with intelligent hierarchical evaluation, precise targeted feedback, and iterative effect traceability. First, an analytic hierarchy process (AHP) is used to establish a multi-dimensional verification index weight model. Each dimension index is weighted and scored, and the verification level is divided into excellent (≥90 points), qualified (70-89 points), and unqualified (<70 points) based on the scoring results. For unqualified indicators, the source of the problem is precisely located, such as voltage sag deviation exceeding the standard corresponding to the load time-varying impedance parameter, and energy storage lifetime deviation exceeding the standard corresponding to the energy storage model attenuation coefficient. This is then fed back to the corresponding modeling unit (such as the dynamic adaptive optimization unit or the new energy and energy storage model integration unit), and targeted correction suggestions are output (such as adjusting the RLC parameter time series coefficient and optimizing the energy storage black-box model parameters). The change curves of the verification indicators before and after each parameter correction are recorded, and a correction effect evaluation report is generated to ensure the effectiveness of the closed-loop correction. Simultaneously, the verification level is linked to the report generation unit. Excellent level reports can be directly used for engineering review, qualified level reports require supplementary local optimization explanations, and unqualified level reports are prohibited from entering the engineering transformation stage.

[0145] In this technical solution, compared with pure software offline simulation, the present invention incorporates actual engineering factors such as communication delay and protocol parsing into the simulation closed loop, effectively verifying the real performance of the controller in regulating wind, solar and storage power output to smooth and support under operating conditions such as load and new energy fluctuations and system failures, providing a basis for on-site parameter tuning.

[0146] Example 4.

[0147] This embodiment 4 discloses a real-time simulation method for electromagnetic transients of source-grid-load-storage coordinated control, wherein step 4 includes the following steps.

[0148] 4.1 The communication configuration unit connects the simulation computing server, energy management controller, wind farm energy management controller, photovoltaic energy management controller, energy storage energy management controller, display, and network switch into a local area network. It sets the server IP (e.g., 192.168.1.10), controller IP (e.g., 192.168.1.XX) and communication port number, and uses built-in tools to test network connectivity.

[0149] 4.2 The virtual communication mapping unit constructs an internal variable and protocol message mapping mechanism, defines the correspondence between simulation internal variables (PCC voltage, current, energy storage SOC, etc.) and communication variables of the energy management controller, wind farm energy management controller, photovoltaic energy management controller, and energy storage energy management controller; and calls the IEC104 protocol stack to realize the forward packaging and reverse unpacking of simulation data and control commands.

[0150] 4.3 The anti-interference processing unit uses the AES-128 encryption algorithm to encrypt the transmitted data and adds a CRC-32 redundant check code; it is configured with dual-network redundant links for primary and backup networks, monitors the primary network latency in real time, and automatically switches to the backup network within 50 milliseconds when the primary network latency exceeds 10ms or a fault occurs.

[0151] 4.4 The delay adaptive compensation unit constructs a communication delay prediction model based on the ARIMA algorithm to predict the delay trend of the next 10 simulation steps; it dynamically adjusts the execution time of the control command according to the prediction results, sets a 50ms delay threshold, and triggers the emergency control logic simulation verification when the threshold is exceeded.

[0152] 4.5 The data interaction unit configures the data transmission frequency, enables the data caching function to avoid data loss caused by instantaneous fluctuations, and synchronously records the data interaction log, including transmission time, content, and status.

[0153] 4.6 Communication Link Accuracy Verification Steps: Test the bit error rate of the communication link (≤10). -12 ), latency fluctuation (≤5ms), data transmission integrity (no packet loss); if the verification fails, return to step 4.3 to optimize anti-interference measures or adjust network parameters; if it passes, proceed to step 5.

[0154] This technical solution employs virtual communication interface technology to eliminate the need for physical interfaces. By inputting commands through the modeling software interface or terminal, a virtual communication mapping module is constructed and point tables are defined. Electrical quantity data is directly extracted from the memory of the simulation computing server and encapsulated into standard communication messages (such as IEC 104) at the software level, then sent directly through the PC's network port. This avoids dependence on physical interface boxes or dedicated servers, significantly reducing the construction cost and maintenance complexity of the experimental platform.

[0155] Example 5.

[0156] This embodiment discloses a real-time simulation method for electromagnetic transients in source-grid-load-storage coordinated control, wherein step 5 includes the following steps.

[0157] 5.1 The control strategy configuration unit installs coordinated control logic such as maximizing wind and solar power absorption, peak-valley arbitrage of energy storage, and load impact suppression on the energy management controller; wind power output regulation strategy on the wind farm energy management controller; photovoltaic maximum power tracking and output smoothing control strategies on the photovoltaic energy management controller; and energy storage charge and discharge control and SOC balancing control strategies on the energy storage energy management controller. It also presets key control parameters for each controller, such as energy storage SOC charge and discharge threshold and power regulation dead zone.

[0158] 5.2 Test Scenario Generation Unit calls the scenario library to select / build basic test scenarios (new energy output fluctuation, load step response, electricity price switching) and extreme operating condition scenarios (new energy sudden drop, multiple loads impacting simultaneously, communication interruption, etc.), and presets the assessment indicators for each scenario.

[0159] 5.3 The real-time simulation operation unit starts the closed-loop linkage between the simulation calculation server and the energy management controller, wind farm energy management controller, photovoltaic energy management controller, and energy storage energy management controller. It constructs a closed-loop link for simulation data acquisition, calculation of commands from each controller, model adjustment, and data feedback. It drives the simulation operation according to the preset scenario and coordinates the simulation step size and the communication frequency of each link to ensure real-time performance.

[0160] 5.4 Dynamic monitoring unit operation: Real-time acquisition of system voltage, current, power, energy storage SOC, controller adjustment commands, etc., and displaying the changing trend in the form of dynamic curves; when the indicators exceed the preset threshold, the abnormal event is automatically marked and the relevant data is recorded.

[0161] Example 6.

[0162] This embodiment discloses a real-time simulation method for electromagnetic transients in coordinated source-grid-load-storage control, wherein step 6 includes the following steps.

[0163] 6.1 The parameter optimization unit adopts the NSGA-Ⅲ multi-objective optimization algorithm, with the objectives of voltage stability, optimal economy, and longest energy storage life. It optimizes the power distribution coefficient and PID adjustment parameters of the controller, the pitch adjustment parameters of the wind farm energy management controller, the MPPT adjustment parameters of the photovoltaic energy management controller, and the charging and discharging threshold of the energy storage energy management controller, and generates a configuration file that can be directly imported into the controller.

[0164] 6.2 The fault risk analysis unit identifies potential fault risks in the system based on simulation data, such as voltage sag nodes or abnormal operating conditions of the energy storage SOC, and outputs early warning thresholds and prevention and control measures. Specifically, it includes the following steps.

[0165] 6.2.1 Risk Data Preprocessing: The simulation data is cleaned and features are extracted to construct a risk analysis dataset. Feature extraction includes extracting 12 types of risk features, such as voltage mutation rate, energy storage SOC fluctuation amplitude, and sudden drop in new energy output. The fault risk analysis unit directly calls the Class A quality data from the data acquisition and preprocessing module, and only performs supplementary analysis on trend anomalies (such as slow parameter drift and periodic fluctuations) during the simulation process.

[0166] 6.2.2 Identification of multiple types of risks.

[0167] 6.2.2.1 Voltage risk identification.

[0168] A. Key node screening: Extract all bus nodes from the simulation model, and prioritize four types of highly sensitive nodes, namely, new energy grid connection points, load concentration access points, transmission line connection nodes, and transformer high-voltage side nodes, to form a voltage risk monitoring node list.

[0169] B. Timing data acquisition: retrieve the voltage timing data of the above nodes under normal and extreme operating conditions, keep the sampling interval consistent with the simulation step size, and extract core features such as voltage amplitude and rate of change.

[0170] C. Risk Trigger Judgment: Compare and analyze the voltage data with the preset trigger threshold. If any of the following conditions are met, it is judged as a voltage risk scenario: a) When the output of new energy sources drops by ≥30%, the node voltage amplitude is ≤0.95pu or ≥1.05pu; b) When the load increases by ≥50%, the node voltage sag is ≥5% and the duration is >0.5s; c) The voltage change rate is >0.05pu / s.

[0171] D. Preliminary risk level assessment: Based on the magnitude and duration of the voltage deviation from the rated value, the risk level is preliminarily determined (e.g., if the voltage is ≤0.9pu and lasts for >2s, it is preliminarily assessed as a level one risk).

[0172] 6.2.2.2 Energy storage risk identification.

[0173] A. Energy storage operation data extraction: Collect data such as SOC value, charging and discharging power, and charging and discharging state switching records of the energy storage system during the entire simulation cycle, and generate energy storage operation sequence curves.

[0174] B. Based on the energy storage model parameters and real-time simulation data determined in step 2 of the modeling stage, conduct SOC anomaly risk identification, and synchronously feed the identification results back to the dynamic adaptive optimization unit to realize the linkage between simulation, risk identification and parameter correction.

[0175] C. Overcharge and over-discharge risk threshold calibration: Based on the safe operation parameters provided by the energy storage equipment manufacturer, and combined with simulation data, the overcharge and over-discharge risk thresholds are deduced. For lithium battery energy storage, SOC ≤ 20% (over-discharge threshold) or ≥ 98% (overcharge threshold), and for vanadium redox flow battery, SOC ≤ 15% (over-discharge threshold) or ≥ 95% (overcharge threshold).

[0176] D. Risk Scenario Recording: For identified abnormal operating conditions, record the trigger time, corresponding simulation operating conditions, and energy storage operation parameters to form an energy storage risk scenario file.

[0177] 6.2.2.3 Identification of system stability risks.

[0178] A. Monitoring of core system indicators: Focus on collecting time-series data of global indicators such as system frequency, total active power, and total reactive power, covering extreme simulation conditions such as sudden drop in new energy output, simultaneous impact of multiple loads, and equipment failure.

[0179] B. Frequency instability judgment: If the system frequency fluctuation is > ±0.2Hz and the duration is > 1s, or the frequency change rate is > 0.1Hz / s, it is judged as a frequency instability risk scenario; if the frequency is ≤ 49.5Hz or ≥ 50.5Hz (power frequency 50Hz scenario), it is directly judged as a level one frequency risk.

[0180] C. Power imbalance judgment: Calculate the total active power deficit of the system. If the deficit is ≥10% of the rated capacity and the duration is >2s, it is judged as a power imbalance risk; if the deficit is ≥20% of the rated capacity, it is marked as an extreme power imbalance scenario.

[0181] D. Risk Correlation Analysis: Analyze the correlation between frequency instability, power imbalance and other risks, such as whether power imbalance is caused by a sudden drop in the output of new energy sources, identify the root cause of the risk, and provide a basis for the formulation of subsequent prevention and control measures.

[0182] 6.2.3 Risk Level Classification: Based on the scope and severity of the risk impact, it is divided into three levels: Level 1 Risk (Critical): May lead to system shutdown and equipment damage, such as overcharging and over-discharging of energy storage, voltage collapse; Level 2 Risk (Severe): Affects the normal operation of the system, such as renewable energy curtailment rate > 5%, voltage fluctuation exceeding the standard; Level 3 Risk (General): Slightly affects economics, such as energy storage revenue lower than expected, small-scale load fluctuation.

[0183] 6.2.4 Early Warning Thresholds and Control Measures: Quantitative early warning thresholds are output for different levels of risk, such as Level 1 risk warning: Energy storage SOC ≤ 25% or ≥ 98%, critical node voltage ≤ 0.9 pu or ≥ 1.1 pu; Simultaneously, targeted control measures are formulated, such as: 1) Voltage sag control: Add SVG reactive power compensation devices at critical nodes and configure a rapid voltage support strategy; 2) Energy storage SOC anomaly control: Optimize charging and discharging strategies, add SOC buffer zones, and avoid extreme charging and discharging; 3) New energy output sudden drop control: Reserve 10%-15% of energy storage reserve capacity and build an output prediction error compensation mechanism.

[0184] 6.3 The report generation unit automatically generates engineering simulation reports, covering simulation conditions, strategy effects, parameter tuning results, risk warnings, etc., and presents them intuitively in a combination of charts and graphs to meet the requirements of engineering design review.

[0185] 6.4 The data export unit exports simulation curve data, optimization parameter tables, verification reports, and other results in standardized Excel, PDF, and CSV formats for data archiving and engineering documentation.

[0186] Example 7.

[0187] This embodiment discloses a real-time simulation method for electromagnetic transients in coordinated source-grid-load-storage control. A multi-dimensional coupled dynamic modeling unit for source-grid-load-storage constructs a multi-dimensional coupled model of electrical, physical, and environmental aspects. By quantifying the coupling correlation rules of parameters in each dimension, coordinated adaptation of electrical characteristics, equipment physical state, and environmental conditions is achieved. The method includes the following steps.

[0188] 1. Based on digital twin technology, a real-time mapping link is established between the physical entities and simulation models of new energy units, energy storage batteries, and industrial loads. The measured physical state parameters are dynamically fed back to the simulation model, and the electrical parameters are corrected synchronously. At the same time, real-time environmental data is integrated to build a coupled driving mechanism to achieve the synergistic quantification of multi-dimensional influences. Finally, the system's comprehensive operating status, including core indicators such as voltage, frequency, power, and energy storage SOC, is output, providing accurate input for subsequent simulation calculations.

[0189] 2. Establish coupling and association across various dimensions.

[0190] 2.1 Electrical and Physical Coupling: The energy storage battery uses 25°C as the reference temperature. For every 1°C increase in temperature, the internal resistance increases by 0.3%. At the same time, a secondary correction mechanism is introduced to compensate for nonlinear effects (when the temperature deviation exceeds 20°C, the correction coefficient is adjusted to 1.1). For industrial loads (such as electric arc furnaces), the initial furnace temperature is used as the reference. For every 1°C deviation in furnace temperature, the load impedance is dynamically corrected by 0.2% to ensure that the electrical characteristics of the load match the physical operating conditions.

[0191] 2.2 Physical and environmental coupling: The pitch angle of the wind turbine is based on the rated wind speed. When the actual wind speed deviates from the rated wind speed, the pitch angle is corrected according to the exponential law. For every 1 m / s deviation of the wind speed from the rated value, the pitch angle is corrected by 5%, which ensures that the physical attitude of the wind turbine is adapted to the environmental wind speed and improves the accuracy of power output calculation.

[0192] 2.3 Environment-Electrical Coupling: The total output of new energy sources is calculated by combining the weights of wind power and photovoltaic output. Wind power output follows the cubic law of wind speed, while photovoltaic output is linearly related to light intensity. At the same time, an environmental temperature correction mechanism is introduced. For every 1°C deviation of the environmental temperature from the 25°C benchmark, the total output of new energy sources is corrected by 0.4%, which is adapted to the output characteristics under extreme temperature conditions.

[0193] 3. By using quantitative multi-dimensional coupling rules, the fragmentation of single-dimensional modeling is avoided; all correction ratios and coefficients are calibrated by fitting measured data to ensure that the simulation error is ≤3%; it can be directly connected to the digital twin platform to realize real-time interaction between physical and simulation data, accurately reproduce the system response characteristics under extreme environments such as typhoons and cold waves and complex working conditions such as sudden changes in equipment status, solve the problem of existing technologies being detached from physical entities and environmental influences and transient simulation distortion, and significantly improve the adaptability of modeling technology to engineering.

[0194] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A real-time simulation method for electromagnetic transients in source-grid-load-storage coordinated control, characterized in that, Includes the following steps: Collect data from multiple scenarios involving source, grid, load, and storage, and preprocess the data to obtain standardized data; Based on the standardized data, an electromagnetic transient simulation model consistent with actual engineering is constructed to achieve accurate modeling and dynamic adaptive optimization of power grids, new energy sources, energy storage, and loads, and to realistically reproduce the electromagnetic transient characteristics under various operating conditions. The electromagnetic transient simulation model includes an industrial load time-varying equivalent model and a multi-dimensional coupled dynamic model. The multi-dimensional coupled dynamic model provides a basis for correction of the industrial load time-varying equivalent model based on multi-dimensional parameters. Based on the standardized data and electromagnetic transient simulation model, a communication link between the simulation computing server and the energy management controller is constructed. A virtual communication interface module is loaded to realize the mapping and encapsulation of simulation internal variables and standard communication messages. Semi-physical real-time simulation is performed based on the communication link.

2. The source-grid-load-storage coordinated control electromagnetic transient real-time simulation method according to claim 1, characterized in that... The steps involved in constructing an electromagnetic transient simulation model are as follows: Establish a model library for intelligent retrieval, dynamic updating, and knowledge transfer; Based on the project's main wiring diagram, topology information is automatically extracted and a structured description file is generated using a deep learning image recognition algorithm. Combined with actual measured parameters of the equipment, the default parameters of the simulation software are corrected to build the electromagnetic transient level system topology. The topology and parameters are linked and verified by generating parameter disturbance scenarios through Monte Carlo simulation. Construct a multi-dimensional coupled model of electrical, physical, and environmental aspects; A black-box model interface adaptive conversion module is used to identify and convert black-box model interface protocols from different manufacturers into a simulation software compatible format. After completing the electrical connection, the offline simulation and online prediction two-stage parameter tuning are started. Construct a hybrid modeling architecture that combines passive components and data-driven equivalent models; Real-time comparison of electromagnetic transient simulation model output with field measured characteristics, applying voltage stability, frequency safety and energy storage lifetime constraints; Static model validation.

3. The source-grid-load-storage coordinated control electromagnetic transient real-time simulation method according to claim 2, characterized in that... Establish the voltage balance equation for the time-varying circuit based on Kirchhoff's voltage law: In the formula, u(t) is the real-time voltage at the load end; i(t) is the real-time current at the load end; R(t) is the time-varying resistance; and L(t) is the time-varying inductance. The time-varying impedance control model is as follows: ; In the formula, R start The initial resistance at the moment of load startup; The resistance attenuation coefficient; t represents the real-time moment; t0 represents the load start-up moment; R stable The resistance is the resistance when the load is running stably; ΔR is the feature compensation coefficient output by the CNN-LSTM model; L start β is the initial inductance at load start-up; β is the inductance temperature influence coefficient; T(t) is the real-time furnace temperature under load; T0 is the initial furnace temperature under load; T max Rated furnace temperature under load; L stable Inductance for stable load operation; A remaining life model for the energy storage battery is constructed to verify the consistency between the energy storage life degradation characteristics in the simulation model and those of the actual battery. The remaining life model for the energy storage battery is as follows: In the formula, L represents the remaining lifespan of the energy storage battery; L base The reference lifespan of the energy storage battery is given by: e is the natural constant; m is the lifespan degradation coefficient; N cyc The cumulative number of charge-discharge cycles for the energy storage battery; D DoD ξ3 is the depth of charge / discharge influence factor; ξ3 is the temperature influence correction coefficient; ΔT bat This is the difference between the actual temperature and the standard temperature of the energy storage battery.

4. The source-grid-load-storage coordinated control electromagnetic transient real-time simulation method according to claim 2, characterized in that... The improved particle swarm optimization algorithm is used to simultaneously optimize model parameters, ensuring that the model matching degree meets the standard while taking into account system stability and component characteristic consistency. The improved particle swarm optimization algorithm model is as follows: , In the formula, F is the overall fitness value; w1 is the weight coefficient of the model deviation term; δ is the relative deviation between the model output and the measured characteristics; w2 is the weight coefficient of the constraint penalty term; c is the constraint category index, which takes values ​​of 1, 2, and 3, corresponding to voltage stability constraint, frequency safety constraint, and energy storage lifetime constraint, respectively. is the penalty coefficient for the c-th constraint; viol(c) is the degree of violation of the c-th constraint.

5. The source-grid-load-storage coordinated control electromagnetic transient real-time simulation method according to claim 2, characterized in that... Constructing a multi-dimensional coupled model includes the following steps: Based on digital twin technology, a real-time mapping link is established between the physical entities and simulation models of new energy units, energy storage batteries, and industrial loads. The measured physical state parameters are dynamically fed back to the simulation model, and the electrical parameters are corrected synchronously. Real-time environmental data is integrated to construct a coupling driving mechanism, quantify the coupling strength and influence weight between electrical parameters, physical state, and environmental factors, and output the comprehensive system operation status including core indicators. Coupling and correlation in various dimensions, including electrical and physical coupling, physical and environmental coupling, and environmental and electrical coupling; By using quantified multi-dimensional coupling rules, the fragmentation of single-dimensional modeling can be avoided.

6. The source-grid-load-storage coordinated control electromagnetic transient real-time simulation method according to claim 1, characterized in that... It also includes multi-dimensional verification of the electromagnetic transient simulation model; the multi-dimensional accuracy verification of the simulation model includes the following steps: A full-time-domain electrical characteristic verification system for transient, steady-state, and harmonic states is constructed; transient voltage oscillation attenuation characteristics are verified using an improved damping coefficient evaluation model. The improved damping coefficient evaluation model is as follows: ； ； ； In the formula, ξ is the improved damping coefficient; ln is the natural logarithm; A k A represents the k-th adjacent oscillation peak of the transient voltage oscillation curve. k+1 ΔP represents the (k+1)th adjacent peak value of the transient voltage oscillation curve; π is the constant pi; k1 is the weight of the impact of new energy power output fluctuations; ΔP renew P is the power output fluctuation coefficient of the new energy source during the oscillation period. renew,k For the actual output of new energy sources during oscillation, P renew,0 Rated output of new energy; k2 is the load impact intensity correction weight; I * imp I represents the per-unit value of the load inrush current. imp I represents the peak value of the actual load inrush current. rated For verifying the load's rated current and voltage imbalance; Based on the principle of energy conservation, and combined with the coupling characteristics of multiple components in the source-grid-load-storage system, an energy balance verification model is constructed. Evaluate the verification results.

7. The source-grid-load-storage coordinated control electromagnetic transient real-time simulation method according to claim 1, characterized in that... Hardware-in-the-loop real-time simulation includes the following steps: The simulation computing server, energy management controller, and network switch are connected to form a local area network, and network connectivity is tested through built-in tools. Construct an internal variable and protocol message mapping mechanism; implement forward packing and reverse unpacking of simulation data and control commands; The system uses encryption algorithms to encrypt transmitted data and adds redundancy checks. It configures dual-network redundant links and monitors the status of the main network in real time. When the main network latency exceeds the standard or fails, it automatically switches to the backup network within 50 milliseconds to ensure stable communication. Construct a communication delay prediction model to predict the delay trend over the next ten simulation steps; dynamically adjust the execution time of control commands based on the prediction results. Configure the data transmission frequency and synchronously record the data interaction log; Communication link accuracy verification.

8. The source-grid-load-storage coordinated control electromagnetic transient real-time simulation method according to claim 1, characterized in that... It also includes a simulation result engineering transformation module that converts simulation verification results into results that can be directly applied to engineering practice, including the following steps: A multi-objective optimization algorithm is used to generate a configuration file that can be directly imported into the controller; Identifying potential system failure risks based on simulation data; Automatically generate engineering simulation reports; The simulation curve data, optimization parameter table, and verification report are visualized and output.

9. The source-grid-load-storage coordinated control electromagnetic transient real-time simulation method according to claim 2, characterized in that... The black-box model interface adaptive conversion module includes an interface protocol identification subunit, a protocol conversion subunit, an adaptation verification subunit, and an exception handling subunit. Interface protocol identification subunit: Automatically scans and identifies the interface type and protocol specifications of the black-box model; Extract interface parameters; Protocol conversion subunit: Based on the identified black-box model interface protocol, it converts it into an interface format compatible with the simulation platform; Adaptation and verification subunit: After completing the protocol conversion, it automatically performs interface connectivity testing, data transmission accuracy verification, and real-time testing; Anomaly Handling Subunit: For issues such as interface identification failure, data transmission packet loss, and protocol incompatibility, it outputs clear alarm information and provides adaptation suggestions.

10. A real-time simulation system for electromagnetic transients of source-grid-load-storage coordinated control using the source-grid-load-storage coordinated control electromagnetic transient real-time simulation method described in claims 1-9, comprising a data acquisition and preprocessing module, an electromagnetic transient simulation modeling module, a multi-dimensional simulation accuracy verification module, a semi-physical real-time simulation system, and a closed-loop simulation verification module, characterized in that... : Data acquisition and preprocessing module: realizes the acquisition, standardized preprocessing and storage of multi-scenario data of source-grid-load-storage system, including data acquisition unit, data preprocessing unit and data storage unit; Electromagnetic transient simulation modeling module: Constructs an electromagnetic transient simulation model of the source-grid-load-storage system consistent with actual engineering, including basic topology modeling unit, new energy and energy storage model integration unit, load disturbance modeling unit, dynamic adaptive optimization unit, source-grid-load-storage multi-coupling dynamic modeling unit and model library unit; Multi-dimensional simulation accuracy verification module: Verifies the simulation model from multiple dimensions of electrical characteristics, physical processes, and economic performance, including electrical dimension verification unit, physical process verification unit, economic dimension verification unit, and verification result evaluation unit; Hardware-in-the-loop real-time simulation system: Constructs a communication link between the simulation computing server and the energy management controller to realize two-way real-time interaction of simulation data and control commands; includes a communication configuration unit, a virtual communication mapping unit, an anti-interference processing unit, a delay adaptive compensation unit, a precision real-time balance control module, and a data interaction unit; Closed-loop simulation verification module: Constructs a closed-loop linkage system between the simulation model and the energy management controller to verify the effectiveness and reliability of the source-grid-load-storage coordinated control strategy; includes a control strategy configuration unit, a test scenario generation unit, a real-time simulation operation unit, and a dynamic monitoring unit; Simulation Result Engineering Transformation Module: This module transforms simulation verification results into outcomes that can be directly applied to engineering practice. It includes a parameter optimization unit, a fault risk analysis unit, a report generation unit, and a data export unit.

Images