Micro-grid group distributed light storage district active support simulation evaluation method and device

Abstract

This invention discloses a method and apparatus for active support simulation evaluation of distributed photovoltaic-storage substations in microgrid clusters, belonging to the field of simulation evaluation technology. The method includes the following steps: acquiring multi-source data of the distributed photovoltaic-storage substations in the microgrid cluster through multi-source sensing devices and data input technology; building a dynamic simulation model of the microgrid cluster with source-storage-load-grid collaboration; obtaining active support simulation evaluation indicators for the microgrid cluster; setting multi-scenario simulation operating condition parameters for the distributed photovoltaic-storage substations in the microgrid cluster; and then obtaining an overall dynamic simulation model of the distributed photovoltaic-storage substations in the microgrid cluster. This enables the construction of a visual simulation evaluation platform and outputs the final active support simulation evaluation report. This invention, through the close integration of multi-source data fusion technology, source-storage-load-grid collaborative modeling technology, 3D modeling technology, and modern information technology, achieves real-time and comprehensive monitoring of the operating status, load characteristics, power flow, and environmental impact of the distributed photovoltaic-storage substations in the microgrid cluster.

Description

Technical Field

[0001] This invention belongs to the field of simulation evaluation technology, specifically relating to a simulation evaluation method and device for active support of distributed photovoltaic-storage distribution areas in microgrid clusters. Background Technology

[0002] With the development trend of distribution networks with a high proportion of distributed photovoltaic and energy storage penetration, traditional distribution area dispatching and operation modes face the challenge of increased randomness and volatility on both the source and load sides. The decentralized layout of photovoltaic and energy storage equipment makes it difficult to achieve efficient and proactive support for voltage stability, load smoothing, and fault recovery in distribution areas through centralized management and control. Existing microgrid simulation and evaluation methods mostly focus on the dimension of a single distribution area or a single device, resulting in problems such as low integration of multi-source data, disconnect between scenario settings and actual operating conditions, incomplete evaluation index system for support capabilities, and lack of data traceability and engineering applicability of simulation results. For example, they do not fully integrate photovoltaic and energy storage equipment operation data, distribution area load data, etc. The synergistic driving effect of power grid topology data, environmental meteorological data, and historical active support event data is insufficient. The coupling degree between the response characteristics of photovoltaic and energy storage equipment, load elasticity regulation characteristics, and power flow characteristics in model construction is inadequate. The simulation scenarios are mostly static assumptions rather than dynamic deductions based on measured data. The evaluation indicators focus on technical performance while ignoring economic efficiency and practicality. Furthermore, the correlation link between simulation results and original data is broken, making it difficult to guide the optimal scheduling and strategy formulation of photovoltaic and energy storage areas. At the same time, existing simulation devices mostly lack integrated visualization and traceability functions, making it impossible to achieve closed-loop management of data acquisition, model construction, scenario simulation, indicator evaluation, and visualization output.

[0003] Traditional active support simulation evaluation techniques suffer from limitations in data acquisition, single simulation dimensions, rigid evaluation mechanisms, and lack of traceability of results. These limitations make it difficult to adapt to the collaborative operation characteristics of distributed photovoltaic and energy storage substations in microgrid clusters. Therefore, there is an urgent need to propose an active support simulation evaluation method for distributed photovoltaic and energy storage substations that integrates all data, couples multiple scenarios, and provides multi-dimensional evaluation, in order to improve the support capability of microgrid clusters for distribution networks and enhance the engineering application value of simulation evaluation. Summary of the Invention

[0004] The purpose of this invention is to provide a simulation and evaluation method and device for active support of distributed photovoltaic and energy storage areas in microgrid clusters. By closely integrating multi-source data fusion technology, source-storage-load-grid collaborative modeling technology, three-dimensional modeling technology and modern information technology, it realizes real-time and comprehensive monitoring of the operating status, load characteristics, power flow and environmental impact of distributed photovoltaic and energy storage areas in microgrid clusters.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A simulation and evaluation method for active support of distributed photovoltaic and energy storage substations in microgrid clusters includes the following steps: Multi-source sensing devices and data entry technology are used to acquire multi-source data of distributed photovoltaic and energy storage distribution areas in microgrid clusters, including photovoltaic and energy storage equipment operation data, distribution area load data, grid topology data, environmental meteorological data, historical active support event data, and physical layout data, and the acquired data is preprocessed. Based on the operation data of photovoltaic and energy storage equipment, the load data of the distribution area, the grid topology data and the environmental meteorological data, a dynamic simulation model of microgrid group with source-storage-load-grid coordination is built to obtain the simulation evaluation index of active support of microgrid group. Historical active support event data is used as scenario-driven parameters to set multi-scenario simulation parameters for distributed photovoltaic and energy storage areas in microgrid clusters. Based on physical layout data, multi-scenario simulation parameters of distributed photovoltaic-storage substations in microgrid clusters, and a dynamic simulation model of microgrid clusters in collaboration with source-storage-load grids, an overall dynamic simulation model of distributed photovoltaic-storage substations in microgrid clusters is built. By utilizing the overall dynamic simulation model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster, a visual simulation evaluation platform is built to output the final active support simulation evaluation report.

[0006] Preferably, the process of acquiring multi-source data from distributed photovoltaic and energy storage substations in a microgrid cluster and preprocessing the acquired data includes: Multi-source data of distributed photovoltaic and energy storage areas in microgrid clusters are collected through multi-source sensing devices and data entry technology. The operational data for photovoltaic and energy storage equipment includes real-time output power, irradiance received value, module temperature, inverter conversion efficiency, and maximum output power limit of the photovoltaic array within the microgrid cluster, as well as real-time state of charge, charge / discharge efficiency, rated capacity, charge / discharge cutoff voltage, and response delay time of the energy storage system; the load data for the distribution area includes real-time active power, real-time reactive power, power factor, electricity consumption time distribution curve, load mutation threshold, rigid load classification labels, and flexible load classification labels for various types of loads within the distribution area; the grid topology data includes the node number of the microgrid cluster connected to the main grid, the type, length, resistance, and reactance of the lines, the status of switching equipment, the rated capacity of transformers, node voltage, and parameters of interconnection lines between microgrids; the environmental meteorological data includes real-time light intensity, real-time temperature, real-time wind speed, real-time wind direction, real-time precipitation, real-time humidity, and the number of extreme weather warnings in the area where the photovoltaic and energy storage distribution area is located. Historical active support event data includes average irradiance, average temperature, average wind speed, average precipitation, average peak load and energy storage efficiency of energy storage devices under historical normal operating events; load fluctuation values ​​and their duration under historical load fluctuation events; average photovoltaic output fluctuation of photovoltaic and energy storage devices under historical photovoltaic and energy storage output fluctuation events; average response delay time of energy storage system and average ambient irradiance; average abnormal voltage of microgrid groups connected to the main grid under historical grid fault events and its duration and event load recovery time; average irradiance, average temperature, average wind speed, average precipitation, average number of extreme weather occurrences and output attenuation of photovoltaic and energy storage devices under historical extreme weather events. The physical layout data includes nameplate information, latitude and longitude coordinates, altitude coordinates, length, width and height of photovoltaic arrays, energy storage cabinets, inverters and transformers, spacing between energy storage cabinets, inverters and transformers, installation tilt angle and row spacing of photovoltaic panels, installation orientation of inverters, latitude and longitude and altitude of the starting and ending points of the distribution area lines, latitude and longitude coordinates and height of the distribution area towers, model and number of conductors of the distribution area lines, connection diagrams of the distribution area lines to the towers and transformers respectively, latitude and longitude coordinates and markings of load nodes, latitude and longitude coordinates of the connection points between loads and distribution area lines, and model of the lines connected to the distribution area. Data cleaning and standardization are performed on the collected data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, the environmental meteorological data, the historical active support event data, and the physical layout data. Timestamps are assigned to the data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, and the environmental meteorological data, and the assigned timestamps are adjusted to achieve synchronization of the collection time of the data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, and the environmental meteorological data.

[0007] Preferably, the process of building a dynamic simulation model of a microgrid cluster that coordinates source, storage, and load networks, and obtaining simulation evaluation indicators for the active support of the microgrid cluster includes: The dynamic simulation model of microgrid clusters with source-storage-load-grid coordination includes a dynamic response sub-model of photovoltaic and storage equipment, a load elasticity adjustment sub-model, a power grid topology and power flow sub-model, and an environmental impact coupling sub-model. The simulation evaluation indicators for active support of microgrid groups include the integrated support response coefficient of photovoltaic and energy storage, the load elasticity adjustment coefficient, the power flow distribution balance coefficient of the grid, and the environmental disturbance coefficient. Combining particle swarm optimization algorithm and backpropagation neural network, and based on the operating data of photovoltaic and energy storage devices, a dynamic response sub-model of photovoltaic and energy storage devices is built, and the process of obtaining the comprehensive support response coefficient of photovoltaic and energy storage devices is as follows: A1. Use the operating data of the photovoltaic and energy storage equipment as the input variable set and the comprehensive support response coefficient of photovoltaic and energy storage as the output variable set; A2. Design a backpropagation neural network structure, which includes an input layer, a hidden layer, and an output layer. Input variables are input into the input layer, and an S-shaped growth curve function is scheduled as the activation function in the hidden layer. A linear function is selected as the activation function of the output layer, and the output is the integrated support response coefficient of the optical storage system. A3. Using the particle swarm optimization algorithm, initialize the particle swarm, define the fitness function, evaluate the performance of each particle, and update the weights and biases of the backpropagation neural network by adjusting the position of each particle, thereby optimizing the weights and biases in the backpropagation neural network. A4. Divide the input variable set into a training set and a test set. Combine the optimized backpropagation neural network to realize the training and verification of the dynamic response sub-model of the photovoltaic storage device, and obtain the final dynamic response sub-model of the photovoltaic storage device. Combine the current operating data of the photovoltaic storage device to output the comprehensive support response coefficient of the photovoltaic storage device.

[0008] Preferably, based on the load data of the transformer substation, a load elasticity adjustment sub-model is built to obtain the load elasticity adjustment coefficient. The process includes: Using fuzzy C-means clustering, load feature matrix is ​​constructed with transformer area load data as the feature dimension to obtain the clustered transformer area load data and its fuzzy membership degree. A long short-term memory network with an attention mechanism is constructed. The clustered transformer area load data and its fuzzy membership degree are used as the input set for load elasticity adjustment. The load elasticity adjustment sub-model is trained by combining long short-term memory network units and attention mechanism. The nonlinear relationship between the clustered transformer area load data and its fuzzy membership degree, the transformer area load data and the load elasticity adjustment coefficient is learned. The final load elasticity adjustment coefficient is output through the output layer to obtain the final load elasticity adjustment sub-model. The current clustered transformer load data and its fuzzy membership degree are input into the load elasticity adjustment sub-model to obtain the load elasticity adjustment coefficient.

[0009] Preferably, based on power grid topology data, a power flow sub-model of the power grid topology is built to obtain the power flow distribution equilibrium coefficient. The process includes: Using matrix construction techniques based on graph theory and circuit theory, a node admittance matrix is ​​constructed based on power grid topology data; The node admittance matrix is ​​decomposed to obtain the real and imaginary parts. The Newton-Raphson method is used to extract the voltage, voltage phase angle, voltage phase angle difference, and impedance of each line segment from the node admittance matrix. The power balance equation is established using Newton-Raphson power flow calculation, the power flow calculation results are obtained, and the actual transmission power of the line is calculated. Based on the actual transmission power of the lines, a power flow sub-model of the power grid topology is built, and the power flow distribution balance coefficient of the power grid is obtained by combining the current power grid topology data.

[0010] Preferably, based on environmental meteorological data, an environmental impact coupling sub-model is built to obtain the environmental disturbance coefficient. The process includes: Environmental meteorological data were integrated into a random forest dataset and divided into a random forest training set and a random forest test set. By using the random forest algorithm, the random forest training set data is used as input data and the environmental disturbance coefficient is used as output data to learn the nonlinear relationship between environmental meteorological data and environmental disturbance coefficient, and a well-trained environmental impact coupling sub-model is obtained. The random forest test set data is input into the trained environmental impact coupling sub-model. By adjusting the parameters of the environmental impact coupling sub-model, the performance of the environmental impact coupling sub-model is optimized, and the final environmental impact coupling sub-model is obtained. The current environmental meteorological data is input into the final environmental impact coupling sub-model, and the environmental disturbance coefficient is output.

[0011] Preferably, the process of using historical active support event data as scenario-driven parameters to set multi-scenario simulation parameters for distributed photovoltaic-storage distribution areas in microgrid clusters includes: The distributed photovoltaic and energy storage distribution areas in microgrid clusters are subject to multiple scenarios, including normal operation scenarios, load fluctuation scenarios, sudden changes in photovoltaic and energy storage output scenarios, grid fault scenarios, and extreme weather scenarios. Using historical active support event data as scenario-driven parameters, and referring to data from historical active support event data under historical normal operation events, historical load fluctuation events, historical photovoltaic and energy storage output change events, historical grid fault events, and historical extreme weather events, the simulation parameters of multi-scenario operating conditions for distributed photovoltaic and energy storage distribution areas in microgrid clusters are set in sequence.

[0012] Preferably, the process of building an overall dynamic simulation model of a distributed photovoltaic-storage substation in a microgrid cluster includes: Unify the coordinate systems for latitude and longitude data, altitude data, and size data within the physical layout data; Using GIS software, based on latitude and longitude data and altitude data in the physical layout data, a three-dimensional terrain surface of the microgrid distributed photovoltaic and energy storage area is generated. Referring to the connection diagrams of the lines in the area with the poles and transformers, and combined with other physical layout data, the geometric model of the equipment in the three-dimensional terrain surface of the microgrid distributed photovoltaic and energy storage area is performed to generate a three-dimensional visualization model of the microgrid distributed photovoltaic and energy storage area. By utilizing model encapsulation technology, the dynamic simulation model of the microgrid cluster with source-storage-load-grid coordination is modularly encapsulated. Combined with modular modeling technology, the modularly encapsulated dynamic simulation model of the microgrid cluster with source-storage-load-grid coordination is interfaced with the 3D visualization model of the distributed photovoltaic-storage area of ​​the microgrid cluster. Combined with data association technology, the simulation parameters of the distributed photovoltaic-storage area of ​​the microgrid cluster in multiple scenarios are configured as the initialization file of the 3D visualization model of the distributed photovoltaic-storage area of ​​the microgrid cluster, thereby realizing the construction of the overall dynamic simulation model of the distributed photovoltaic-storage area of ​​the microgrid cluster.

[0013] Preferably, the process of building a visual simulation evaluation platform and outputting the final active support simulation evaluation report includes: A real-time simulation engine is deployed to compile the overall dynamic simulation model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster into functional model units that conform to the functional model interface standard. The compiled functional model units are then imported into the real-time simulation engine. The compiled functional model units are called through the functional model interface to output the active support simulation evaluation index of the microgrid cluster. This enables the active support simulation evaluation of the distributed photovoltaic and energy storage area of ​​the microgrid cluster, thereby realizing the construction of a visual simulation evaluation platform and obtaining the active support simulation evaluation results of the distributed photovoltaic and energy storage area of ​​the microgrid cluster. Using a visualization simulation evaluation platform, combined with multi-source data of the current distributed photovoltaic and energy storage areas in the microgrid cluster, we can obtain the active support simulation evaluation results and active support simulation evaluation indicators of the current distributed photovoltaic and energy storage areas in the microgrid cluster. By integrating the current active support simulation evaluation results of distributed photovoltaic and energy storage substations in microgrid clusters, the active support simulation evaluation indicators of microgrid clusters, and multi-source data of distributed photovoltaic and energy storage substations in microgrid clusters, and using data report generation and visualization technology, the final active support simulation evaluation report is output.

[0014] A simulation and evaluation device for active support of distributed photovoltaic and energy storage substations in microgrid clusters, used to implement the above methods, includes: The multi-source data sensing module acquires multi-source data from the distributed photovoltaic and energy storage areas of the microgrid cluster through multi-source sensing devices and data entry technology, and preprocesses the acquired data. The dynamic sub-simulation module, based on the operating data of photovoltaic and energy storage devices, the load data of distribution areas, the grid topology data, and environmental meteorological data, builds a dynamic simulation model of a microgrid cluster that coordinates source, storage, load, and grid, and obtains the simulation evaluation index of the microgrid cluster's active support. The dynamic sub-simulation module includes the following units: The dynamic response unit of the photovoltaic and energy storage equipment is based on the backpropagation neural network of the particle swarm optimization algorithm and the operation data of the photovoltaic and energy storage equipment to build a dynamic response sub-model of the photovoltaic and energy storage equipment and output the comprehensive support response coefficient of photovoltaic and energy storage. The load elasticity adjustment unit is based on the long short-term memory network combined modeling technology of fuzzy C-means clustering and attention mechanism and transformer area load data to build a load elasticity adjustment sub-model and obtain the load elasticity adjustment coefficient. The power grid topology power flow unit is based on the power system topology dynamic reconfiguration optimization algorithm of the Newton-Raphson method and power grid topology data. A power grid topology power flow sub-model is built to obtain the power grid power flow distribution equilibrium coefficient. The environmental impact coupling unit uses the random forest algorithm and environmental meteorological data to build an environmental impact coupling sub-model and obtain the environmental disturbance coefficient. The multi-scenario simulation condition parameter determination module uses historical active support event data as scenario-driven parameters to set multi-scenario simulation condition parameters for the distributed photovoltaic-storage area of ​​the microgrid cluster. The dynamic overall simulation module builds an overall dynamic simulation model of the microgrid cluster distributed photovoltaic-storage area based on physical layout data, multi-scenario simulation parameters of the microgrid cluster distributed photovoltaic-storage area, and the dynamic simulation model of the microgrid cluster coordinated by the source-storage-load network. The active support simulation evaluation and visualization module utilizes the overall dynamic simulation model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster to build a visualization simulation evaluation platform and output the final active support simulation evaluation report.

[0015] The beneficial effects of this invention are: This invention comprehensively integrates six types of data—photovoltaic-storage equipment, transformer area load, power grid topology, environmental meteorology, historical supporting events, and physical layout—through multi-source data acquisition and precise preprocessing. This achieves time-series synchronization and standardized processing, addressing the problems of fragmented data acquisition, low integration, and spatiotemporal asynchrony in traditional methods from the data source. Simultaneously, it constructs four dynamic sub-models for source-storage-load-grid collaboration, employing algorithms such as particle swarm optimization BP neural network, fuzzy C-means clustering with attention mechanism LSTM, Newton-Raphson method, and random forest. These algorithms accurately output four core evaluation indicators: photovoltaic-storage response, load regulation, power flow balance, and environmental interference. This enables deep coupling and quantitative evaluation of multi-dimensional characteristics, significantly improving the accuracy and adaptability of the simulation model.

[0016] This invention is based on historical active support event-driven multi-scenario operating condition settings, covering five typical scenarios: normal operation, load fluctuation, photovoltaic-storage sudden change, grid failure, and extreme weather, making the simulation more closely resemble actual operating conditions. It also combines GIS 3D modeling and modular encapsulation technology to build an overall dynamic simulation model and a visualization evaluation platform, realizing the integrated linkage of 3D spatial layout and dynamic simulation. Finally, it outputs a traceable and visualized complete evaluation report, effectively solving the problems of single scenario, black box results, and lack of engineering applicability of traditional technologies, and significantly improving the intelligence, accuracy, and practicality of microgrid group active support capability assessment. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic block diagram of the device of the present invention; Figure 3 Output curves for the dynamic response sub-model of photovoltaic storage devices; Figure 4 Output curves for the load elastic adjustment sub-model; Figure 5 Output curves for the environmental impact coupled sub-model; Figure 6 A 3D visualization model of a distributed photovoltaic-storage area in a microgrid cluster; Figure 7 Simulation evaluation curves for active support of microgrid groups in multiple scenarios. Detailed Implementation

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0019] Example 1: As Figure 1 As shown, the simulation and evaluation method for active support of distributed photovoltaic and energy storage substations in microgrid clusters includes the following steps: S1. By using multi-source sensing devices and data entry technology, multi-source data of distributed photovoltaic and energy storage areas in microgrid clusters are acquired, and the acquired data is preprocessed to provide a data foundation for subsequent steps, thus solving the problem of data acquisition limitations in traditional active support simulation and evaluation technology. S2. Based on the operation data of photovoltaic and energy storage equipment, the load data of distribution areas, the grid topology data and the environmental meteorological data, a dynamic simulation model of microgrid groups with source-storage-load-grid coordination is built to obtain the simulation evaluation index of active support of microgrid groups, which provides technical and data support for realizing the simulation evaluation of active support of distributed photovoltaic and energy storage distribution areas in microgrid groups. S3. Using historical active support event data as scenario-driven parameters, multi-scenario simulation parameters for the distributed photovoltaic-storage area of ​​the microgrid cluster are set, providing multi-scenario simulation parameters for the subsequent construction of the overall dynamic simulation model of the distributed photovoltaic-storage area of ​​the microgrid cluster. This makes the final active support simulation evaluation results more in line with the actual situation and improves the accuracy of the active support simulation evaluation results. S4. Based on physical layout data, multi-scenario simulation parameters of distributed photovoltaic-storage substations in microgrid clusters, and a dynamic simulation model of microgrid clusters with source-storage-load network collaboration, an overall dynamic simulation model of distributed photovoltaic-storage substations in microgrid clusters is built. This realizes the fusion of the dynamic simulation model of microgrid clusters with source-storage-load network collaboration, and incorporates the actual multi-scenario simulation parameters of distributed photovoltaic-storage substations in microgrid clusters into the reference range of active support simulation evaluation, thereby improving the efficiency of active support simulation evaluation and laying the foundation for the technical implementation of active support simulation evaluation of distributed photovoltaic-storage substations in microgrid clusters. S5. By utilizing the overall dynamic simulation model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster, a visual simulation evaluation platform is built to output the final active support simulation evaluation report. This realizes the visual output of the active support simulation evaluation results of the distributed photovoltaic and energy storage area of ​​the microgrid cluster, and solves the problem of lack of traceability of results in traditional active support simulation evaluation technology.

[0020] In S1, the process of acquiring multi-source data from the distributed photovoltaic and energy storage areas of the microgrid cluster through multi-source sensing devices and data entry technology, and preprocessing the acquired data includes: Multi-source sensing devices and data entry technology are used to collect multi-source data from distributed photovoltaic and energy storage areas in microgrid clusters. The multi-source sensing devices include power meters, irradiance sensors, temperature sensors, DC power sensors, AC power sensors, battery management systems, energy meters, energy storage system controllers, power factor meters, load dispatching systems, load monitoring systems, intelligent switchgear monitoring systems, voltage sensors, light sensors, wind speed sensors, wind direction sensors, rain gauges, humidity sensors, GPS receivers, lidar, laser rangefinders, tilt angle meters, and direction meters. The operational data for photovoltaic and energy storage equipment includes real-time output power, irradiance received value, module temperature, inverter conversion efficiency, and maximum output power limit of the photovoltaic array within the microgrid cluster, as well as real-time state of charge, charge / discharge efficiency, rated capacity, charge / discharge cutoff voltage, and response delay time of the energy storage system; the load data for the distribution area includes real-time active power, real-time reactive power, power factor, electricity consumption time distribution curve, load mutation threshold, rigid load classification labels, and flexible load classification labels for various types of loads within the distribution area; the grid topology data includes the node number of the microgrid cluster connected to the main grid, the type, length, resistance, and reactance of the lines, the status of switching equipment, the rated capacity of transformers, node voltage, and parameters of interconnection lines between microgrids; the environmental meteorological data includes real-time light intensity, real-time temperature, real-time wind speed, real-time wind direction, real-time precipitation, real-time humidity, and the number of extreme weather warnings in the area where the photovoltaic and energy storage distribution area is located. Historical active support event data includes average irradiance, average temperature, average wind speed, average precipitation, average peak load and energy storage efficiency of energy storage devices under historical normal operating events; load fluctuation values ​​and their duration under historical load fluctuation events; average photovoltaic output fluctuation of photovoltaic and energy storage devices under historical photovoltaic and energy storage output fluctuation events; average response delay time of energy storage system and average ambient irradiance; average abnormal voltage of microgrid groups connected to the main grid under historical grid fault events and its duration and event load recovery time; average irradiance, average temperature, average wind speed, average precipitation, average number of extreme weather occurrences and output attenuation of photovoltaic and energy storage devices under historical extreme weather events. The physical layout data includes nameplate information, latitude and longitude coordinates, altitude coordinates, length, width and height of photovoltaic arrays, energy storage cabinets, inverters and transformers, spacing between energy storage cabinets, inverters and transformers, installation tilt angle and row spacing of photovoltaic panels, installation orientation of inverters, latitude and longitude and altitude of the starting and ending points of the distribution area lines, latitude and longitude coordinates and height of the distribution area towers, model and number of conductors of the distribution area lines, connection diagrams of the distribution area lines to the towers and transformers respectively, latitude and longitude coordinates and markings of load nodes, latitude and longitude coordinates of the connection points between loads and distribution area lines, and model of the lines connected to the distribution area. Using a power meter, irradiance sensor, temperature sensor, DC power sensor, and AC power sensor, the real-time output power, irradiance received value, module temperature, and DC input power and AC output power of the photovoltaic array in the microgrid are collected. The inverter conversion efficiency of the photovoltaic array in the microgrid is obtained by the ratio of the AC output power to the DC input power of the inverter. The maximum output power limit of the photovoltaic array in the microgrid is obtained from the energy management system database using data entry technology. Using the battery management system, the real-time state of charge value, rated capacity, and charge / discharge cutoff voltage of the energy storage system are collected. Combined with the energy meter and energy storage system controller, the discharge energy, charging energy, and response delay time of the energy storage system are collected. The charge / discharge efficiency of the energy storage system is obtained by the percentage of discharge energy in the charging energy. Power meters are used to collect real-time active and reactive power data of various loads within the distribution area. Power factor meters are used to collect the power factor data of various loads within the distribution area. Through the load dispatching system, the change data of various loads are recorded in real time. Combined with timestamps, the power consumption time distribution curves of various loads within the distribution area are generated. Through the load monitoring system, the changes of various loads within the distribution area are tracked in real time to obtain the load change threshold of various loads within the distribution area. Data entry technology is used to obtain the rigid load classification labels and flexible load classification labels of various loads within the distribution area from the power demand response management system database. Using data entry technology, the node number, line type, and transformer rated capacity of the microgrid group connected to the main grid are obtained from the microgrid management system and transformer nameplate information, respectively. The model, length, resistance, and reactance of the lines connecting the power grid group to the main grid, as well as the parameters of the interconnection lines between microgrids, are obtained from the power grid design archive database. The status of the switching equipment of the microgrid group connected to the main grid is collected through the intelligent switching equipment monitoring system, and the node voltage of the microgrid group connected to the main grid is collected using voltage sensors. Using light sensors, temperature sensors, wind speed sensors, wind direction sensors, rain gauges, and humidity sensors, the real-time light intensity, real-time temperature, real-time wind speed, real-time wind direction, real-time precipitation, and real-time humidity of the area where the photovoltaic storage station is located are obtained. Using data entry technology, the number of extreme weather warnings in the area where the photovoltaic storage station is located within the past 24 hours is obtained from the meteorological early warning system. Based on data entry technology, for historical routine operation events, average irradiance, average temperature, average wind speed, and average precipitation are obtained from the meteorological bureau's data platform, and average peak load and energy storage efficiency of energy storage devices are extracted from the power grid company's operation database. For historical load fluctuation events, load mutation values ​​and their durations are obtained through the power dispatch system. For historical photovoltaic-energy storage output mutation events, the average photovoltaic output mutation value of photovoltaic-energy storage devices and the average response delay time of the energy storage system are obtained from the photovoltaic power generation monitoring system, and the average ambient irradiance is obtained from the meteorological bureau's data platform. For historical power grid fault events, the average abnormal voltage of microgrid groups connected to the main grid and its duration are obtained from the power grid fault database, and the event load recovery time is obtained from the microgrid control system. For historical extreme weather events, the average irradiance, average temperature, average wind speed, average precipitation, and average number of extreme weather occurrences are obtained from the extreme weather event database, and the output attenuation of photovoltaic-energy storage devices is selected from the photovoltaic data platform. By combining a GPS receiver, lidar, laser rangefinder, tilt meter, and direction meter, the system collects the latitude, longitude, altitude, length, width, and height of photovoltaic arrays, energy storage cabinets, inverters, and transformers; the spacing between energy storage cabinets, inverters, and transformers; the installation tilt angle and row spacing of photovoltaic panels; the installation orientation of inverters; the latitude, longitude, and altitude of the starting and ending points of the distribution area lines; the latitude, longitude, and altitude of the distribution area towers; the latitude, longitude, and latitude coordinates of load nodes; and the latitude, longitude, and latitude coordinates of the connection points between loads and distribution area lines. Through data entry technology, the system extracts the nameplate information of photovoltaic arrays, energy storage cabinets, inverters, and transformers from the manufacturer databases. It obtains the model and conductor split number of the distribution area lines from the power equipment management system. Through the GIS system, it queries the connection relationships between the distribution area lines and towers and transformers, thereby obtaining connection diagrams of the distribution area lines to towers and transformers respectively. It obtains the identification of load nodes and the model of the distribution area lines connected to the connection points between loads and distribution area lines from the power load management system. Data cleaning and standardization are performed on the collected data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, the environmental meteorological data, the historical active support event data, and the physical layout data to eliminate outliers and interference from different parameters. Timestamps are assigned to the data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, and the environmental meteorological data, and the timestamps are adjusted to achieve synchronization of the collection time of the data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, and the environmental meteorological data.

[0021] In S2, based on the operating data of photovoltaic and energy storage devices, the load data of distribution areas, the grid topology data, and the environmental meteorological data, a dynamic simulation model of a microgrid group coordinating source, storage, load, and grid is built. The process of obtaining the simulation evaluation index of the microgrid group's active support includes: The dynamic simulation model of microgrid clusters with source-storage-load-grid coordination includes a dynamic response sub-model of photovoltaic and storage equipment, a load elasticity adjustment sub-model, a power grid topology and power flow sub-model, and an environmental impact coupling sub-model. The simulation evaluation indicators for active support of microgrid groups include the integrated support response coefficient of photovoltaic and energy storage, the load elasticity adjustment coefficient, the power flow distribution balance coefficient of the grid, and the environmental disturbance coefficient. Combining particle swarm optimization algorithm and backpropagation neural network, and based on the operating data of photovoltaic and energy storage devices, a dynamic response sub-model of photovoltaic and energy storage devices is built, and the process of obtaining the comprehensive support response coefficient of photovoltaic and energy storage devices is as follows: A1. Use the operating data of the photovoltaic and energy storage equipment as the input variable set and the comprehensive support response coefficient of photovoltaic and energy storage as the output variable set; A2. Design a backpropagation neural network structure, which includes an input layer, a hidden layer, and an output layer. Input variables are input into the input layer. In the hidden layer, the sigmoid function is used as the activation function to achieve nonlinear mapping of the features of the input variable set. A linear function is selected as the activation function of the output layer to output the integrated support response coefficient of the optical storage system. A3. Using the particle swarm optimization algorithm, initialize the particle swarm. Each particle represents a set of weights and biases of the backpropagation neural network. Define a fitness function to evaluate the performance of each particle. This fitness function is used to minimize the error of the backpropagation neural network. By adjusting the position of each particle, update the weights and biases of the backpropagation neural network to optimize the weights and biases in the backpropagation neural network, thereby improving the accuracy of the integrated support response coefficient of the optical storage system output by the backpropagation neural network. A4. Divide the input variable set into a training set and a test set in a 7:3 ratio. Combine the optimized backpropagation neural network to train and verify the dynamic response sub-model of the photovoltaic storage device, and obtain the final dynamic response sub-model of the photovoltaic storage device. Combine the current operating data of the photovoltaic storage device to output the comprehensive support response coefficient of the photovoltaic storage device.

[0022] In S2, the process of building a load elasticity adjustment sub-model and obtaining the load elasticity adjustment coefficient using transformer area load data includes: Using fuzzy C-means clustering, a load feature matrix is ​​constructed with transformer area load data as the feature dimension. The clustered transformer area load data and their fuzzy membership degrees are obtained, and a fuzzy membership degree threshold is set. Rigid load classification labels and flexible load classification labels are used as supervisory constraints, and the number of clusters is set to 2. These two clusters include rigid clusters and flexible clusters. The real-time active power, real-time reactive power, power factor, electricity consumption time distribution curve, and load mutation threshold of various loads within the transformer area are classified into load characteristics within each cluster. The objective function is iteratively optimized to minimize the transformer area load characteristics within each cluster. The fuzzy membership degree of each transformer load sample is obtained by summing the squared fuzzy errors of the features. This represents the probability that the transformer load sample belongs to a rigid load or an elastic load. Transformer load samples with fuzzy membership degrees lower than the fuzzy membership degree threshold are then removed. Finally, the typical transformer load features within the rigid and elastic clusters are retained. The transformer load data after the above clustering are the typical transformer load features within the rigid and elastic clusters. The fuzzy membership degree of the transformer load data after the above clustering is the fuzzy membership degree of the typical transformer load features within the rigid and elastic clusters. A long short-term memory network (LSTM) with an attention mechanism is constructed to learn the temporal variation patterns and elastic characteristics of transformer area load. Clustered transformer area load data and their fuzzy membership degrees are used as the input set for load elasticity adjustment, enabling parallel input of multi-dimensional data. The load elasticity adjustment sub-model is trained by combining LSM network units and the attention mechanism, learning the nonlinear relationship between clustered transformer area load data and their fuzzy membership degrees, and between transformer area load data and load elasticity adjustment coefficients. The final load elasticity adjustment coefficients are output through the output layer, resulting in the final load elasticity adjustment sub-model. The LSM network units capture the long-term temporal dependencies between clustered transformer area load data and their fuzzy membership degrees. The attention mechanism allows the LSM network units to assign corresponding weights to the load elasticity adjustment input set at each time step according to different transformer area load types, ensuring that elastic load features are prioritized. The output layer uses a mean squared error loss function combined with a gradient descent algorithm to iteratively optimize the load elasticity adjustment sub-model parameters until the load elasticity adjustment sub-model converges, obtaining the final load elasticity adjustment sub-model. The current clustered transformer load data and its fuzzy membership degree are input into the load elasticity adjustment sub-model to obtain the load elasticity adjustment coefficient.

[0023] In S2, the process of building a power flow sub-model based on power grid topology data and obtaining the power flow distribution equilibrium coefficient includes: Using matrix construction techniques based on graph theory and circuit theory, a node admittance matrix is ​​constructed based on power grid topology data. This node admittance matrix is ​​a core step in power system power flow calculation. It is a complex matrix that reflects the electrical coupling relationship between nodes in the power grid. The node admittance matrix is ​​decomposed to obtain its real and imaginary parts. The Newton-Raphson method is used to extract the voltage, voltage phase angle, voltage phase angle difference, and impedance of each line segment from the node admittance matrix. The extraction process can be implemented by programming with the numpy library in Python. The power balance equation is established using Newton-Raphson power flow calculation, the power flow calculation results are obtained, and the actual transmission power of the line is calculated. Based on the actual transmission power of the line, a power flow sub-model of the power grid topology is built, and the power flow distribution balance coefficient of the power grid is obtained by combining the current power grid topology data. After converting the voltage and voltage phase angle difference at each node, as well as the real and imaginary parts of the node admittance matrix, into dimensionless per-unit values, the power balance equation is as follows: ; in, and Let be the injected active power and injected reactive power of node i, respectively. and Let be the per-unit values ​​of the voltages at nodes i and j, respectively. and , respectively, are the per-unit values ​​of the real and imaginary parts of the nodal admittance matrix. Let n be the phase angle difference between node i and node j. Node i refers to the target node being studied, and node j is all the adjacent nodes directly connected to node i. Therefore, n is the total number of all adjacent nodes directly connected to node i. The formulas for calculating the actual transmission power of a line include: ; ; ; ; ; ; in, Let be the branch current phasor of the line between node i and node j. and Let be the voltage phasors from node i to node j, respectively. Let be the impedance of the line between node i and node j. and Let be the voltages at nodes i and j, respectively. and Let be the voltage phase angles at nodes i and j, respectively. Let be the active power flowing from node i to node j. Let be the active power flowing from node j to node i. for The conjugate phasor, To obtain the real part, This represents the actual power transmitted by the line. yes The reverse power, as shown in the formula for calculating the actual transmission power of the line, is... and The maximum value obtained after taking the modulo is... ; The expression for the power flow sub-model of the power grid topology is as follows: ; in, This is the power flow distribution equilibrium coefficient of the power grid. This represents the actual power transmitted by the line. Determined by the line model, it represents the line's rated transmission power. The standard deviation is represented by the power flow distribution balance coefficient. The larger the coefficient, the smaller the difference in load rate among the lines and the more balanced the power flow distribution. Its value ranges from 0 to 1.

[0024] In S2, the process of building an environmental impact coupling sub-model based on environmental meteorological data and obtaining the environmental disturbance coefficient includes: Environmental meteorological data were integrated into a random forest dataset and divided into a random forest training set and a random forest test set in an 8:2 ratio. By using the random forest algorithm, the random forest training set data is used as input data and the environmental disturbance coefficient is used as output data to learn the nonlinear relationship between environmental meteorological data and environmental disturbance coefficient, and a well-trained environmental impact coupling sub-model is obtained. The random forest test set data is input into the trained environmental impact coupling sub-model. By adjusting the parameters of the environmental impact coupling sub-model, the performance of the environmental impact coupling sub-model is optimized, and the final environmental impact coupling sub-model is obtained. The current environmental meteorological data is input into the final environmental impact coupling sub-model, and the environmental disturbance coefficient is output.

[0025] In S3, the process of setting the simulation parameters for multiple scenarios of distributed photovoltaic-storage distribution areas in microgrid clusters by using historical active support event data as scenario-driven parameters includes: The distributed photovoltaic and energy storage distribution areas in microgrid clusters are subject to multiple scenarios, including normal operation scenarios, load fluctuation scenarios, sudden changes in photovoltaic and energy storage output scenarios, grid fault scenarios, and extreme weather scenarios. Using historical active support event data as scenario-driven parameters, and referring to the data in the historical active support event data under historical normal operation events, historical load fluctuation events, historical photovoltaic and energy storage output change events, historical power grid fault events and historical extreme weather events, the simulation parameters of multi-scenario operating conditions of microgrid group distributed photovoltaic and energy storage area are set in turn. The average light intensity, average temperature, average wind speed, average precipitation, average peak load, and energy storage efficiency of the energy storage device under historical normal operating events are respectively set as the critical values ​​for light intensity, temperature, wind speed, precipitation, peak load, and energy storage efficiency of the distributed photovoltaic energy storage area of ​​the microgrid group under normal operating scenarios. The load mutation value and its duration under historical load fluctuation events are set as the threshold values ​​for load mutation and duration of distributed photovoltaic energy storage areas in microgrid clusters under load fluctuation scenarios, respectively. The average photovoltaic output mutation value of photovoltaic and energy storage equipment, the average response delay time of energy storage system, and the average ambient light intensity under historical photovoltaic and energy storage output mutation events are respectively set as the photovoltaic output mutation threshold, the response delay time threshold, and the ambient light intensity threshold of photovoltaic and energy storage equipment in the distributed photovoltaic and energy storage area of ​​microgrid group under photovoltaic and energy storage output mutation scenario. The average abnormal voltage and its duration, along with the event load recovery time, of the microgrid clusters connected to the main grid under historical grid fault events are set as the thresholds for the abnormal voltage and its duration of the microgrid clusters connected to the main grid under grid fault scenarios, and the event load recovery time threshold, respectively. The average light intensity, average temperature, average wind speed, average precipitation, average number of extreme weather events, and power attenuation of photovoltaic and energy storage devices under historical extreme weather events are respectively set as the critical values ​​for light intensity, temperature, wind speed, precipitation, number of extreme weather events, and power attenuation of photovoltaic and energy storage devices in the distributed photovoltaic and energy storage area of ​​the microgrid cluster under extreme weather scenarios.

[0026] In S4, the process of building the overall dynamic simulation model of the distributed photovoltaic-storage area of ​​the microgrid cluster, based on physical layout data, multi-scenario simulation parameters of the distributed photovoltaic-storage area of ​​the microgrid cluster, and the dynamic simulation model of the source-storage-load network in coordination with the microgrid cluster, includes: The coordinate system of geospatial data, including latitude and longitude data, altitude data, and size data, is unified within the physical layout data. Among them, the latitude and longitude data in the physical layout data includes the latitude and longitude of photovoltaic arrays, energy storage cabinets, inverters, and transformers, the latitude and longitude of the starting and ending points of the distribution area lines, the latitude and longitude coordinates of the distribution area towers, and the latitude and longitude coordinates of the connection points between the load and the distribution area lines. The altitude data in the physical layout data includes the altitude of photovoltaic arrays, energy storage cabinets, inverters, and transformers, as well as the altitude of the starting and ending points of the distribution area lines. The size data in the physical layout data includes the length, width, and height of photovoltaic arrays, energy storage cabinets, inverters, and transformers, as well as the height of the distribution area towers. Using GIS software, based on latitude and longitude data and altitude data in the physical layout data, a three-dimensional terrain surface of the microgrid distributed photovoltaic and energy storage area is generated. Referring to the connection diagrams of the lines in the area with the poles and transformers, and combined with other physical layout data, the geometric model of the equipment in the three-dimensional terrain surface of the microgrid distributed photovoltaic and energy storage area is performed to generate a three-dimensional visualization model of the microgrid distributed photovoltaic and energy storage area. By utilizing model encapsulation technology, the dynamic simulation model of a microgrid cluster integrating source, storage, load, and grid is modularly encapsulated. Combined with modular modeling technology, the modularly encapsulated dynamic simulation model of the microgrid cluster is interfaced with the 3D visualization model of the distributed photovoltaic-storage area within the microgrid cluster. Each dynamic simulation model of the microgrid cluster integrates source, storage, load, and grid through a data interface to achieve real-time data interaction, ensuring the timeliness of the overall dynamic simulation model of the distributed photovoltaic-storage area. This enables online updates for subsequent active support simulation evaluation of the distributed photovoltaic-storage area. Combined with data association technology, the simulation parameters of the distributed photovoltaic-storage area in multiple scenarios are configured into the initialization file of the 3D visualization model of the distributed photovoltaic-storage area. This provides a basis for subsequently driving the overall dynamic simulation model of the distributed photovoltaic-storage area under different preset scenarios, thus realizing the construction of the overall dynamic simulation model of the distributed photovoltaic-storage area.

[0027] In S5, the process of building a visual simulation evaluation platform and outputting the final active support simulation evaluation report by utilizing the overall dynamic simulation model of the distributed photovoltaic-storage area of ​​the microgrid cluster includes: A real-time simulation engine compatible with the Functional Model Interface (FMI) is deployed to compile the overall dynamic simulation model of the distributed photovoltaic-storage area of ​​the microgrid cluster into functional model units that conform to the FMI standard. The compiled functional model units are then imported into the real-time simulation engine. By calling the compiled functional model units through the FMI, the active support simulation evaluation indicators of the microgrid cluster are output. This enables the active support simulation evaluation of the distributed photovoltaic-storage area of ​​the microgrid cluster, realizing the construction of a visual simulation evaluation platform and obtaining the active support simulation evaluation results of the distributed photovoltaic-storage area of ​​the microgrid cluster. This active support simulation evaluation of the distributed photovoltaic-storage area of ​​the microgrid cluster incorporates multi-scenario simulation operating condition parameters and physical layout factors of the distributed photovoltaic-storage area of ​​the microgrid cluster, improving the accuracy and real-world fit of the active support simulation evaluation of the distributed photovoltaic-storage area of ​​the microgrid cluster in practical applications. Using a visualization simulation evaluation platform, combined with multi-source data of the current distributed photovoltaic and energy storage areas in the microgrid cluster, we can obtain the active support simulation evaluation results and active support simulation evaluation indicators of the current distributed photovoltaic and energy storage areas in the microgrid cluster. By integrating the current active support simulation evaluation results, active support simulation evaluation indicators, and multi-source data of distributed photovoltaic and energy storage areas in microgrid clusters, and employing data report generation and visualization technologies, a final active support simulation evaluation report is output. Through three-level traceability of indicator data, active support simulation evaluation results, and original multi-source data, each active support simulation evaluation result can be traced back to specific original multi-source data and indicator data. This solves the black box problem in traditional active support simulation evaluation technology, making it more operable for subsequent operators to intervene in distributed photovoltaic and energy storage areas of microgrid clusters.

[0028] The algorithm involved in this embodiment can be executed by an electronic device, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The processor executes the software to implement the above-mentioned algorithm calculation.

[0029] Example 2: Figure 2 As shown, the active support simulation and evaluation device for distributed photovoltaic and energy storage areas in a microgrid cluster is used to implement the method in the embodiment. It includes a multi-source data sensing module, a dynamic sub-simulation module, a multi-scenario simulation condition parameter determination module, a dynamic overall simulation module, and an active support simulation evaluation and visualization module. The modules are interconnected. The multi-source data sensing module acquires multi-source data from distributed photovoltaic and energy storage areas in microgrid clusters through multi-source sensing devices and data input technology, and preprocesses the acquired data, thus solving the problem of data acquisition limitations in traditional active support simulation and evaluation technologies. The dynamic sub-simulation module, based on the operation data of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data and the environmental meteorological data, builds a dynamic simulation model of the microgrid group with source, storage and load network coordination, and obtains the active support simulation evaluation index of the microgrid group. The dynamic sub-simulation module is divided into a dynamic response unit of photovoltaic and energy storage equipment, a load elastic adjustment unit, a power grid topology flow unit and an environmental impact coupling unit, which solves the problem of the single simulation dimension in traditional active support simulation evaluation technology. The dynamic response unit of the photovoltaic and energy storage equipment is based on the backpropagation neural network of the particle swarm optimization algorithm and the operation data of the photovoltaic and energy storage equipment to build a dynamic response sub-model of the photovoltaic and energy storage equipment and output the comprehensive support response coefficient of photovoltaic and energy storage. The load elasticity adjustment unit is based on the long short-term memory network combined modeling technology of fuzzy C-means clustering and attention mechanism and transformer area load data to build a load elasticity adjustment sub-model and obtain the load elasticity adjustment coefficient. The power grid topology power flow unit is based on the power system topology dynamic reconfiguration optimization algorithm of the Newton-Raphson method and power grid topology data. A power grid topology power flow sub-model is built to obtain the power grid power flow distribution equilibrium coefficient. The environmental impact coupling unit uses the random forest algorithm and environmental meteorological data to build an environmental impact coupling sub-model and obtain the environmental disturbance coefficient. The multi-scenario simulation condition parameter determination module uses historical active support event data as scenario-driven parameters to set multi-scenario simulation condition parameters for the distributed photovoltaic-storage area of ​​the microgrid cluster. The dynamic overall simulation module, based on physical layout data, multi-scenario simulation parameters of distributed photovoltaic-storage substations in microgrid clusters, and a dynamic simulation model of microgrid clusters in collaboration with the source-storage-load network, builds an overall dynamic simulation model of distributed photovoltaic-storage substations in microgrid clusters, solving the problem of rigid evaluation mechanism in traditional active support simulation evaluation technology; The active support simulation evaluation and visualization module utilizes the overall dynamic simulation model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster to build a visualization simulation evaluation platform and output the final active support simulation evaluation report.

[0030] The implementation details of each module are the same as in Example 1.

[0031] Example 3: This example is a verification illustration of Example 1, combined with... Figures 3-7 The effectiveness of the method in Example 1 was verified: Figure 3 The output curve of the dynamic response sub-model of the photovoltaic and energy storage device is used to show the dynamic response performance of the photovoltaic and energy storage device at different times. Figure 3 The horizontal axis represents time in hours, ranging from 0 to 24 hours, while the vertical axis represents the integrated support response coefficient of photovoltaic and energy storage, ranging from 0 to 1. A higher value of the integrated support response coefficient indicates a stronger active support response capability of the photovoltaic and energy storage equipment. This dynamic response sub-model of photovoltaic and energy storage equipment is built based on particle swarm optimization algorithm and backpropagation neural network, which intuitively reflects the changes in the response characteristics of photovoltaic and energy storage equipment during the simulation period, providing key data support for the simulation evaluation of active support of microgrid groups.

[0032] Figure 4 It is the output curve of the load elasticity adjustment sub-model, which is used to show the changes in the elasticity adjustment capability of the transformer area load at different times. Figure 4 The horizontal axis represents time in hours, ranging from 0 to 24 hours, while the vertical axis represents the load flexibility coefficient, ranging from 0 to 1. A higher load flexibility coefficient indicates a stronger load flexibility capability and greater adaptability to the operational needs of the microgrid group. The load flexibility sub-model is generated by a combination of fuzzy C-means clustering and attention mechanism-based long short-term memory network modeling techniques, trained based on distribution area load data. It intuitively reflects the dynamic changes in load flexibility characteristics and is one of the important data references for the active support simulation evaluation of microgrid groups.

[0033] Figure 5It is the output curve of the environmental impact coupling sub-model, used to show the change of the degree of interference of environmental factors on the operation of microgrid groups over time; Figure 5 The horizontal axis in the model represents time in hours, ranging from 0 to 24 hours. The vertical axis represents the environmental interference coefficient, ranging from 0 to 1. A higher environmental interference coefficient indicates stronger environmental interference on the active support of the microgrid cluster. The environmental impact coupling sub-model is generated by modeling using the random forest algorithm and trained based on environmental meteorological data. It intuitively reflects the dynamic changes in environmental interference characteristics and provides key data references for the environmental dimension of the simulation evaluation of the active support of the microgrid cluster.

[0034] Figure 6 This is a 3D visualization model of a distributed photovoltaic and energy storage distribution area in a microgrid cluster, used to intuitively present the geographical spatial layout and distribution of core equipment in the area. The 3D visualization model of the distributed photovoltaic and energy storage distribution area in the microgrid cluster is based on a 3D coordinate system. Figure 6 The horizontal axis represents longitude, the vertical axis represents latitude, and the elevation axis represents altitude, ranging from 44 to 56 meters, clearly reproducing the elevation differences in the plateau area. Figure 6 The model clearly marks key elements such as terrain, photovoltaic arrays, energy storage cabinets, inverters, transformers, towers, and typical loads, completely replicating the core equipment layout and geographical relationships of the distributed photovoltaic and energy storage area of ​​the microgrid cluster. The 3D visualization model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster is obtained by generating a 3D terrain surface and performing equipment geometric modeling based on physical layout data through GIS software. It is an important component of the overall dynamic simulation model of the microgrid cluster and provides an intuitive spatial carrier for subsequent multi-scenario simulation and visualization evaluation.

[0035] Figure 7 This is a simulation evaluation curve for active support of microgrid groups in multiple scenarios, used to intuitively present the differences in the impact of different operating scenarios on the active support capability of microgrid groups. Figure 7 The horizontal axis represents time in hours, ranging from 0 to 24 hours. The vertical axis represents the active support simulation evaluation index, corresponding to the microgrid group's active support simulation evaluation index, with a value ranging from 0 to 1. A higher value for the microgrid group's active support simulation evaluation index indicates stronger active support capability. Figure 7 The curves in the figure represent five typical scenarios: normal operation, load fluctuation, sudden changes in photovoltaic and energy storage output, grid failure, and extreme weather. Each scenario corresponds to a curve, which clearly compares the dynamic changes in the active support capability of microgrid groups under different scenarios. The simulation operating parameters of these scenarios are set based on historical active support event data. This figure provides direct data support for the optimization scheduling and strategy formulation of microgrid groups under different actual operating scenarios.

[0036] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A simulation and evaluation method for active support of distributed photovoltaic-storage distribution areas in microgrid clusters, characterized in that, Includes the following steps: Multi-source sensing devices and data entry technology are used to acquire multi-source data of distributed photovoltaic and energy storage distribution areas in microgrid clusters, including photovoltaic and energy storage equipment operation data, distribution area load data, grid topology data, environmental meteorological data, historical active support event data, and physical layout data, and the acquired data is preprocessed. Based on the operation data of photovoltaic and energy storage equipment, the load data of the distribution area, the grid topology data and the environmental meteorological data, a dynamic simulation model of microgrid group with source-storage-load-grid coordination is built to obtain the simulation evaluation index of active support of microgrid group. Historical active support event data is used as scenario-driven parameters to set multi-scenario simulation parameters for distributed photovoltaic and energy storage areas in microgrid clusters. Based on physical layout data, multi-scenario simulation parameters of distributed photovoltaic-storage substations in microgrid clusters, and a dynamic simulation model of microgrid clusters in collaboration with source-storage-load grids, an overall dynamic simulation model of distributed photovoltaic-storage substations in microgrid clusters is built. By utilizing the overall dynamic simulation model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster, a visual simulation evaluation platform is built to output the final active support simulation evaluation report.

2. The simulation and evaluation method for active support of distributed photovoltaic-storage substations in microgrid clusters according to claim 1, characterized in that, The process of acquiring multi-source data from distributed photovoltaic and energy storage distribution areas in a microgrid cluster and preprocessing the acquired data includes: Multi-source data of distributed photovoltaic and energy storage areas in microgrid clusters are collected through multi-source sensing devices and data entry technology. The operational data for photovoltaic and energy storage equipment includes real-time output power, irradiance received value, module temperature, inverter conversion efficiency, and maximum output power limit of the photovoltaic array within the microgrid cluster, as well as real-time state of charge, charge / discharge efficiency, rated capacity, charge / discharge cutoff voltage, and response delay time of the energy storage system; the load data for the distribution area includes real-time active power, real-time reactive power, power factor, electricity consumption time distribution curve, load mutation threshold, rigid load classification labels, and flexible load classification labels for various types of loads within the distribution area; the grid topology data includes the node number of the microgrid cluster connected to the main grid, the type, length, resistance, and reactance of the lines, the status of switching equipment, the rated capacity of transformers, node voltage, and parameters of interconnection lines between microgrids; the environmental meteorological data includes real-time light intensity, real-time temperature, real-time wind speed, real-time wind direction, real-time precipitation, real-time humidity, and the number of extreme weather warnings in the area where the photovoltaic and energy storage distribution area is located. Historical active support event data includes average irradiance, average temperature, average wind speed, average precipitation, average peak load and energy storage efficiency of energy storage devices under historical normal operating events; load fluctuation values ​​and their duration under historical load fluctuation events; average photovoltaic output fluctuation of photovoltaic and energy storage devices under historical photovoltaic and energy storage output fluctuation events; average response delay time of energy storage system and average ambient irradiance; average abnormal voltage of microgrid groups connected to the main grid under historical grid fault events and its duration and event load recovery time; average irradiance, average temperature, average wind speed, average precipitation, average number of extreme weather occurrences and output attenuation of photovoltaic and energy storage devices under historical extreme weather events. The physical layout data includes nameplate information, latitude and longitude coordinates, altitude coordinates, length, width and height of photovoltaic arrays, energy storage cabinets, inverters and transformers, spacing between energy storage cabinets, inverters and transformers, installation tilt angle and row spacing of photovoltaic panels, installation orientation of inverters, latitude and longitude and altitude of the starting and ending points of the distribution area lines, latitude and longitude coordinates and height of the distribution area towers, model and number of conductors of the distribution area lines, connection diagrams of the distribution area lines to the towers and transformers respectively, latitude and longitude coordinates and markings of load nodes, latitude and longitude coordinates of the connection points between loads and distribution area lines, and model of the lines connected to the distribution area. Data cleaning and standardization are performed on the collected data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, the environmental meteorological data, the historical active support event data, and the physical layout data. Timestamps are assigned to the data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, and the environmental meteorological data, and the assigned timestamps are adjusted to achieve synchronization of the collection time of the data on the operation of photovoltaic and energy storage equipment, the load data of the distribution area, the power grid topology data, and the environmental meteorological data.

3. The simulation and evaluation method for active support of distributed photovoltaic-storage substations in microgrid clusters according to claim 1, characterized in that, The process of building a dynamic simulation model of a microgrid cluster with coordinated source, storage, and load networks, and obtaining simulation evaluation indicators for the active support of the microgrid cluster includes: The dynamic simulation model of microgrid clusters with source-storage-load-grid coordination includes a dynamic response sub-model of photovoltaic and storage equipment, a load elasticity adjustment sub-model, a power grid topology and power flow sub-model, and an environmental impact coupling sub-model. The simulation evaluation indicators for active support of microgrid groups include the integrated support response coefficient of photovoltaic and energy storage, the load elasticity adjustment coefficient, the power flow distribution balance coefficient of the grid, and the environmental disturbance coefficient. Combining particle swarm optimization algorithm and backpropagation neural network, and based on the operating data of photovoltaic and energy storage devices, a dynamic response sub-model of photovoltaic and energy storage devices is built, and the process of obtaining the comprehensive support response coefficient of photovoltaic and energy storage devices is as follows: A1. Use the operating data of the photovoltaic and energy storage equipment as the input variable set and the comprehensive support response coefficient of photovoltaic and energy storage as the output variable set; A2. Design a backpropagation neural network structure, which includes an input layer, a hidden layer, and an output layer. Input variables are input into the input layer, and an S-shaped growth curve function is scheduled as the activation function in the hidden layer. A linear function is selected as the activation function of the output layer, and the output is the integrated support response coefficient of the optical storage system. A3. Using the particle swarm optimization algorithm, initialize the particle swarm, define the fitness function, evaluate the performance of each particle, and update the weights and biases of the backpropagation neural network by adjusting the position of each particle, thereby optimizing the weights and biases in the backpropagation neural network. A4. Divide the input variable set into a training set and a test set. Combine the optimized backpropagation neural network to realize the training and verification of the dynamic response sub-model of the photovoltaic storage device, and obtain the final dynamic response sub-model of the photovoltaic storage device. Combine the current operating data of the photovoltaic storage device to output the comprehensive support response coefficient of the photovoltaic storage device.

4. The simulation and evaluation method for active support of distributed photovoltaic-storage substations in microgrid clusters according to claim 3, characterized in that, Based on the load data of the transformer substation, a load elasticity adjustment sub-model is built to obtain the load elasticity adjustment coefficient. The process includes: Using fuzzy C-means clustering, load feature matrix is ​​constructed with transformer area load data as the feature dimension to obtain the clustered transformer area load data and its fuzzy membership degree. A long short-term memory network with an attention mechanism is constructed. The clustered transformer area load data and its fuzzy membership degree are used as the input set for load elasticity adjustment. The load elasticity adjustment sub-model is trained by combining long short-term memory network units and attention mechanism. The nonlinear relationship between the clustered transformer area load data and its fuzzy membership degree, the transformer area load data and the load elasticity adjustment coefficient is learned. The final load elasticity adjustment coefficient is output through the output layer to obtain the final load elasticity adjustment sub-model. The current clustered transformer load data and its fuzzy membership degree are input into the load elasticity adjustment sub-model to obtain the load elasticity adjustment coefficient.

5. The simulation and evaluation method for active support of distributed photovoltaic-storage substations in microgrid clusters according to claim 3, characterized in that, Based on power grid topology data, a power flow sub-model of the power grid topology is constructed to obtain the power flow distribution equilibrium coefficient. The process includes: Using matrix construction techniques based on graph theory and circuit theory, a node admittance matrix is ​​constructed based on power grid topology data; The node admittance matrix is ​​decomposed to obtain the real and imaginary parts. The Newton-Raphson method is used to extract the voltage, voltage phase angle, voltage phase angle difference, and impedance of each line segment from the node admittance matrix. The power balance equation is established using Newton-Raphson power flow calculation, the power flow calculation results are obtained, and the actual transmission power of the line is calculated. Based on the actual transmission power of the lines, a power flow sub-model of the power grid topology is built, and the power flow distribution balance coefficient of the power grid is obtained by combining the current power grid topology data.

6. The simulation and evaluation method for active support of distributed photovoltaic-storage substations in microgrid clusters according to claim 3, characterized in that, Based on environmental meteorological data, an environmental impact coupling sub-model was constructed to obtain the environmental disturbance coefficient. The process includes: Environmental meteorological data were integrated into a random forest dataset and divided into a random forest training set and a random forest test set. By using the random forest algorithm, the random forest training set data is used as input data and the environmental disturbance coefficient is used as output data to learn the nonlinear relationship between environmental meteorological data and environmental disturbance coefficient, and a well-trained environmental impact coupling sub-model is obtained. The random forest test set data is input into the trained environmental impact coupling sub-model. By adjusting the parameters of the environmental impact coupling sub-model, the performance of the environmental impact coupling sub-model is optimized, and the final environmental impact coupling sub-model is obtained. The current environmental meteorological data is input into the final environmental impact coupling sub-model, and the environmental disturbance coefficient is output.

7. The simulation and evaluation method for active support of distributed photovoltaic-storage substations in microgrid clusters according to claim 2, characterized in that, The process of using historical active support event data as scenario-driven parameters to set multi-scenario simulation parameters for distributed photovoltaic and energy storage substations in microgrid clusters includes: The distributed photovoltaic and energy storage distribution areas in microgrid clusters are subject to multiple scenarios, including normal operation scenarios, load fluctuation scenarios, sudden changes in photovoltaic and energy storage output scenarios, grid fault scenarios, and extreme weather scenarios. Using historical active support event data as scenario-driven parameters, and referring to data from historical active support event data under historical normal operation events, historical load fluctuation events, historical photovoltaic and energy storage output change events, historical grid fault events, and historical extreme weather events, the simulation parameters of multi-scenario operating conditions for distributed photovoltaic and energy storage distribution areas in microgrid clusters are set in sequence.

8. The simulation and evaluation method for active support of distributed photovoltaic-storage substations in microgrid clusters according to claim 2, characterized in that, The process of building a dynamic simulation model of the distributed photovoltaic and energy storage area of ​​a microgrid cluster includes: Unify the coordinate systems for latitude and longitude data, altitude data, and size data within the physical layout data; Using GIS software, based on latitude and longitude data and altitude data in the physical layout data, a three-dimensional terrain surface of the microgrid distributed photovoltaic and energy storage area is generated. Referring to the connection diagrams of the lines in the area with the poles and transformers, and combined with other physical layout data, the geometric model of the equipment in the three-dimensional terrain surface of the microgrid distributed photovoltaic and energy storage area is performed to generate a three-dimensional visualization model of the microgrid distributed photovoltaic and energy storage area. By utilizing model encapsulation technology, the dynamic simulation model of the microgrid cluster with source-storage-load-grid coordination is modularly encapsulated. Combined with modular modeling technology, the modularly encapsulated dynamic simulation model of the microgrid cluster with source-storage-load-grid coordination is interfaced with the 3D visualization model of the distributed photovoltaic-storage area of ​​the microgrid cluster. Combined with data association technology, the simulation parameters of the distributed photovoltaic-storage area of ​​the microgrid cluster in multiple scenarios are configured as the initialization file of the 3D visualization model of the distributed photovoltaic-storage area of ​​the microgrid cluster, thereby realizing the construction of the overall dynamic simulation model of the distributed photovoltaic-storage area of ​​the microgrid cluster.

9. The simulation and evaluation method for active support of distributed photovoltaic-storage substations in microgrid clusters according to claim 1, characterized in that, The process of building a visual simulation evaluation platform and outputting the final active support simulation evaluation report includes: A real-time simulation engine is deployed to compile the overall dynamic simulation model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster into functional model units that conform to the functional model interface standard. The compiled functional model units are then imported into the real-time simulation engine. The compiled functional model units are called through the functional model interface to output the active support simulation evaluation index of the microgrid cluster. This enables the active support simulation evaluation of the distributed photovoltaic and energy storage area of ​​the microgrid cluster, thereby realizing the construction of a visual simulation evaluation platform and obtaining the active support simulation evaluation results of the distributed photovoltaic and energy storage area of ​​the microgrid cluster. Using a visualization simulation evaluation platform, combined with multi-source data of the current distributed photovoltaic and energy storage areas in the microgrid cluster, we can obtain the active support simulation evaluation results and active support simulation evaluation indicators of the current distributed photovoltaic and energy storage areas in the microgrid cluster. By integrating the current active support simulation evaluation results of distributed photovoltaic and energy storage substations in microgrid clusters, the active support simulation evaluation indicators of microgrid clusters, and multi-source data of distributed photovoltaic and energy storage substations in microgrid clusters, and using data report generation and visualization technology, the final active support simulation evaluation report is output.

10. A simulation and evaluation device for active support of distributed photovoltaic-storage substations in microgrid clusters, used to implement the method described in any one of claims 1-9, characterized in that, include: The multi-source data sensing module acquires multi-source data from the distributed photovoltaic and energy storage areas of the microgrid cluster through multi-source sensing devices and data entry technology, and preprocesses the acquired data. The dynamic sub-simulation module, based on the operating data of photovoltaic and energy storage devices, the load data of distribution areas, the grid topology data, and environmental meteorological data, builds a dynamic simulation model of a microgrid cluster that coordinates source, storage, load, and grid, and obtains the simulation evaluation index of the microgrid cluster's active support. The dynamic sub-simulation module includes the following units: The dynamic response unit of the photovoltaic and energy storage equipment is based on the backpropagation neural network of the particle swarm optimization algorithm and the operation data of the photovoltaic and energy storage equipment to build a dynamic response sub-model of the photovoltaic and energy storage equipment and output the comprehensive support response coefficient of photovoltaic and energy storage. The load elasticity adjustment unit is based on the long short-term memory network combined modeling technology of fuzzy C-means clustering and attention mechanism and transformer area load data to build a load elasticity adjustment sub-model and obtain the load elasticity adjustment coefficient. The power grid topology power flow unit is based on the power system topology dynamic reconfiguration optimization algorithm of the Newton-Raphson method and power grid topology data. A power grid topology power flow sub-model is built to obtain the power grid power flow distribution equilibrium coefficient. The environmental impact coupling unit uses the random forest algorithm and environmental meteorological data to build an environmental impact coupling sub-model and obtain the environmental disturbance coefficient. The multi-scenario simulation condition parameter determination module uses historical active support event data as scenario-driven parameters to set multi-scenario simulation condition parameters for the distributed photovoltaic-storage area of ​​the microgrid cluster. The dynamic overall simulation module builds an overall dynamic simulation model of the microgrid cluster distributed photovoltaic-storage area based on physical layout data, multi-scenario simulation parameters of the microgrid cluster distributed photovoltaic-storage area, and the dynamic simulation model of the microgrid cluster coordinated by the source-storage-load network. The active support simulation evaluation and visualization module utilizes the overall dynamic simulation model of the distributed photovoltaic and energy storage area of ​​the microgrid cluster to build a visualization simulation evaluation platform and output the final active support simulation evaluation report.

Images