Virtual power plant and microgrid collaborative scheduling system based on multi-objective optimization

Abstract

The application discloses a virtual power plant and micro-grid collaborative scheduling system based on multi-objective optimization, and particularly relates to the field of multi-objective collaborative scheduling, and comprises the following steps: a data acquisition module acquires and standardizes distributed new energy output, load, energy storage, power grid parameters and market carbon price information; a multi-objective optimization module constructs economic cost, carbon emission, power supply reliability and new energy consumption sub-targets, adopts an analytic hierarchy process and a dynamic fuzzy membership degree to adaptively correct weights, couples multiple targets into a comprehensive optimization target, and solves a global optimal scheduling strategy; a collaborative scheduling module converts the strategy into an instruction and issues the instruction, monitors in real time and starts an emergency collaborative mechanism; and an execution feedback module completes instruction execution and data return, forming a closed-loop iteration; the application realizes multi-objective collaborative optimization and adaptive scheduling, improves system economy, reliability and new energy consumption capacity, and is suitable for efficient and stable operation of the virtual power plant and the micro-grid.

Description

Technical Field

[0001] This invention relates to the field of multi-objective collaborative scheduling technology, and more specifically, to a virtual power plant and microgrid collaborative scheduling system based on multi-objective optimization. Background Technology

[0002] With the rapid development of new energy power generation technologies, distributed new energy sources are being connected to the grid on a large scale. Virtual power plants and microgrids, as core carriers for integrating distributed energy and improving energy utilization efficiency, are playing a crucial role in the coordinated dispatch of power system operations and management. Currently, the power system is transforming towards low-carbon, intelligent, and market-oriented directions. Users are increasingly demanding higher power supply reliability and new energy absorption capacity. At the same time, the gradual improvement of the electricity market and carbon trading market is also placing higher requirements on the coordinated operation of virtual power plants and microgrids. Virtual power plants aggregate multiple microgrids and distributed energy resources to achieve coordinated allocation and optimized utilization of energy. Microgrids, on the other hand, play a crucial role in local energy supply and load regulation. The coordinated dispatch of these two systems directly affects the safe and stable operation of the power system and the improvement of its overall efficiency. Against this backdrop, how to achieve efficient coordination between virtual power plants and microgrids, rationally allocate various energy resources, and adapt to multi-dimensional operational needs has become a critical issue that urgently needs to be addressed in the current power system field. This also drives the continuous exploration and development of technologies related to the coordinated dispatch of virtual power plants and microgrids.

[0003] However, it still has some drawbacks in practical use, such as: 1. Existing virtual power plant and microgrid dispatching technologies mostly adopt single-objective optimization, which makes it difficult to take into account multiple needs such as economy, low carbon, power supply reliability, and new energy consumption. There is a lack of reasonable weight allocation among the objectives, and the dispatching results are poorly adapted to the actual operation scenario, which cannot meet the comprehensive management and control requirements of complex power grids. 2. Traditional scheduling methods rely on fixed weight settings and cannot adaptively adjust according to the real-time status of the power grid. When faced with scenarios such as fluctuations in new energy output, load changes, and equipment malfunctions, the response flexibility is insufficient, which can easily lead to resource waste or a decline in power supply security. 3. Existing technologies mostly adopt passive adjustment strategies in handling anomalies, lack a unified emergency coordination mechanism, make it difficult to quickly call up backup capacity and reconfigure power flow, and have a slow fault recovery speed, which affects the safe and stable operation of the system and the power supply guarantee for critical loads. 4. Most scheduling systems have not formed a complete closed-loop iterative mechanism. Execution results and running data cannot be effectively fed back to the optimization stage, making it difficult to continuously correct and upgrade scheduling strategies. Long-term operating accuracy gradually decreases, and the level of intelligence and adaptability is limited. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art, the present invention provides a virtual power plant and microgrid collaborative scheduling system based on multi-objective optimization, which solves the problems mentioned in the background art through the following scheme.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization, comprising: Data acquisition module: performs real-time, multi-dimensional data acquisition, anomaly cleaning, missing data completion, and dimensional unification processing of distributed renewable energy output, electricity load, energy storage operation status, grid node parameters, and electricity market and carbon price information to form standardized time-series data; Multi-objective optimization module: Based on standardized time series data, four sub-objectives are constructed: economic cost, carbon emissions, power supply reliability, and new energy consumption. Static basic weights are determined by the analytic hierarchy process and dynamic fuzzy membership is used to adaptively correct the weights in combination with the real-time operation status of the power grid. Multiple objectives are strongly coupled into a single comprehensive optimization objective. The global optimal scheduling strategy is obtained under the constraints of equipment and power grid safety. The collaborative scheduling module parses the globally optimal scheduling strategy into standardized scheduling instructions, which are then sent to each microgrid and terminal device through dual redundant communication channels. It monitors line power, energy storage status, and reserve capacity constraints in real time. In cases of exceeding limits or failures, it dynamically adjusts the scheduling instructions according to the priority of comprehensive objectives and activates a safety-first emergency collaborative mechanism. Execution feedback module: used to drive new energy, energy storage, load and transmission equipment to execute scheduling commands, collect the actual operating status and command execution deviation in real time, and synchronously transmit the operating data and verification results back to the data acquisition unit and multi-objective optimization unit to form a closed-loop iterative operation system.

[0006] The technical effects and advantages of this invention are as follows: 1. This solution constructs a multi-objective system that integrates economy, low carbon emissions, reliable power supply, and renewable energy consumption. Through a unified framework, it achieves multi-objective coupling optimization, which can fully match the multiple needs of power grid operation and improve the comprehensive applicability and rationality of the dispatching scheme. 2. By combining static weights with dynamic adaptive correction, the target priority is dynamically adjusted based on the real-time operating status, which greatly improves the flexibility of the scheduling strategy and can effectively adapt to complex scenarios such as new energy fluctuations and load changes. 3. Establish a safety-first emergency coordination mechanism, which can quickly identify anomalies and proactively allocate resources, enabling rapid power reconfiguration and precise redeployment of backup capacity, significantly improving fault handling speed and system operational stability; 4. Establish a closed-loop iterative system covering the entire process from data collection, optimization, scheduling to execution feedback. Real-time data transmission and continuous strategy correction enable continuous improvement in scheduling accuracy and significantly enhanced system intelligence and self-optimization capabilities. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of the overall structure of the present invention.

[0008] Figure 2 This is a schematic diagram illustrating the interaction process between the data acquisition module and the multi-objective optimization module of the present invention.

[0009] Figure 3 This is a schematic diagram of the interaction process between the multi-objective optimization module and the collaborative scheduling module of the present invention.

[0010] Figure 4 This is a schematic diagram of the interaction process between the collaborative scheduling module and the execution feedback module of the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] refer to Figure 1 - Figure 4 The virtual power plant and microgrid collaborative dispatch system shown is based on multi-objective optimization and includes: Data acquisition module: Collects data from the virtual power plant and various microgrids during operation, including new energy output, user load, energy storage status, grid operation parameters and market environment data. It performs data cleaning, anomaly detection, interpolation and standardization, and outputs a time-series dataset in a unified format to provide data input for the multi-objective optimization module and the collaborative scheduling module.

[0013] It should be further explained that the data acquisition module includes: a data acquisition terminal, a data transmission unit, a data preprocessing unit, and a data storage unit; It should be further explained that the data acquisition terminals are deployed in the virtual power plant control center and various microgrid nodes, including new energy output acquisition terminals, load acquisition terminals, energy storage acquisition terminals, grid status acquisition terminals, and market data acquisition terminals. All terminals adopt a unified acquisition protocol to ensure data format consistency; the details are as follows: The new energy power generation data acquisition terminal is responsible for collecting real-time active power of photovoltaic power generation, real-time reactive power of photovoltaic power generation, irradiance of photovoltaic array, temperature of photovoltaic modules, real-time active power of wind turbine, real-time reactive power of wind turbine, real-time wind speed, real-time wind direction, wind turbine speed, operating status of new energy power generation equipment, and fault status of new energy power generation equipment. The load acquisition terminal is responsible for collecting real-time power of residential electricity load, real-time power of commercial electricity load, real-time power of industrial electricity load, real-time power of adjustable load, real-time power of non-adjustable load, real-time power of total electricity load, load power factor, load voltage, load current, load operating period, and load type identifier. The energy storage data acquisition terminal is responsible for collecting real-time state of charge (SOC) of the energy storage battery pack, real-time charging power of the energy storage, real-time discharging power of the energy storage, charging current of the energy storage, discharging current of the energy storage, energy storage terminal voltage, energy storage battery temperature, energy storage charging and discharging efficiency, energy storage rated capacity, maximum allowable charging power of the energy storage, maximum allowable discharging power of the energy storage, energy storage equipment operating status, and energy storage equipment fault status. The power grid status acquisition terminal is responsible for collecting the voltage amplitude at the point of common coupling (PCC), the current amplitude at the PCC, the real-time frequency of the system, the real-time active power transmitted by the lines, the real-time reactive power transmitted by the lines, the transmission current of the lines, the node voltage deviation, the active power loss of the lines, the reactive power loss of the lines, the three-phase imbalance of the power grid, the harmonic distortion rate of the power grid, the operating status of the power grid, and the fault signals of the power grid. The market data acquisition terminal is responsible for collecting real-time spot electricity prices, time-of-use spot electricity prices, on-grid electricity prices, electricity purchase prices, real-time carbon prices in the carbon trading market, ancillary service market prices, demand response subsidy prices, and peak-valley time period identifiers for the power grid.

[0014] It should be further explained that the data transmission unit adopts a 5G / fiber dual-mode transmission method and is equipped with a data encryption module (AES-256 encryption) to prevent data leakage or tampering; the transmission latency is ≤50ms. This parameter is determined based on the overall real-time requirements of the system—the total latency from data acquisition to optimization solution must be ≤100ms, and 50ms is allocated as a data transmission reserve time. Combining the actual performance of 5G (latency ≤10ms) and fiber (latency ≤5ms), the reserved redundancy ensures transmission stability, and the communication success rate is ≥99.9%.

[0015] It should be further explained that the data preprocessing unit incorporates data cleaning, anomaly detection, interpolation completion, and standardization algorithms. Its core function is to eliminate data noise and unify units, laying the foundation for multi-objective coupled optimization, as detailed below: Data cleaning: Remove data that exceeds a reasonable range (such as data where the output of new energy exceeds the rated power of the equipment by ±5%) and abnormally fluctuating data (such as data where the load fluctuates by more than 20% within 1 minute). Anomaly detection: Thresholding method and isolated forest algorithm are used. The threshold is determined by the equipment's rated parameters and historical operating data statistics. The training samples for the isolated forest algorithm are historical data from the past 3 months. The detection accuracy is ≥98%. Interpolation completion: For missing data (missing rate ≤ 5%), linear interpolation is used for completion; when the missing rate is > 5%, LSTM model prediction is used for completion to ensure data integrity. Standardization: The min-max normalization method is used to unify all data to the [0, 1] interval, eliminating dimensional differences (such as the difference between kW-level power output and kV-level voltage), providing a unified scale basis for subsequent AHP weight calculation and target coupling. The standardization formula is: ;in, The data is standardized, and x represents the original data. Add 5% redundancy to the historical minimum value for this type of data. The standardization is adjusted by adding 5% redundancy to the historical maximum value of this type of data. This redundancy is used to prevent extreme data from causing standardization failure. and Updated every 7 days based on the latest historical data.

[0016] Further explanation is needed regarding the data storage unit: it adopts a hybrid storage mode of edge computing + cloud backup. The edge stores high-frequency data collected in real time (sampling frequency: 1 min / time for new energy output and load data, 500 ms / time for grid status data, and 15 min / time for market data), while the cloud stores historical data, pre-processed data, and optimized scheduling results. The sampling frequency is determined based on the data fluctuation characteristics: fluctuations in new energy and load are at the minute level, fluctuations in grid status are at the millisecond level, and the market data update cycle is 15-30 min. This setting balances real-time performance with energy consumption and storage pressure. The storage capacity meets the requirement of retaining at least one year of historical data, and is determined according to the power system dispatch data retention specifications (1-3 years) and the actual data volume (approximately 500 GB of data per year). It supports data query, traceability, and retrieval.

[0017] It should be further explained that the workflow of the data acquisition module is as follows: Data Acquisition Startup: After the system is powered on, each acquisition terminal automatically starts up and collects various types of raw data according to the preset sampling frequency. The acquisition terminal performs a self-check periodically (once every 10 minutes) and immediately reports any abnormalities to the data transmission unit. Data transmission: The acquisition terminal transmits the raw data to the data preprocessing unit in real time through the encrypted 5G / fiber optic transmission unit. If data loss occurs during transmission, a retransmission mechanism is automatically triggered (retransmission times ≤ 3 times) to ensure uninterrupted data transmission. Data preprocessing: The preprocessing unit processes the raw data in the order of "cleaning → anomaly detection → interpolation completion → standardization", and outputs a standardized time series dataset in a unified format. The total preprocessing delay is ≤50ms, ensuring that the data is delivered to the multi-objective optimization module in a timely manner. Data storage and output: The preprocessed standardized data is output in real time to the multi-objective optimization module and the collaborative scheduling module for real-time optimization and scheduling; and is also stored synchronously at the edge and cloud to form a historical database, providing data support for AHP weight calibration and constraint parameter adjustment of the multi-objective optimization module.

[0018] Multi-objective optimization module: Based on the standardized data from the data acquisition module, four sub-objective functions and a unified constraint set are constructed. The AHP+dynamic fuzzy membership method is used to achieve strong coupling of sub-objectives and construct a unique comprehensive optimization objective. An improved adaptive optimization algorithm is used to solve the global optimal scheduling scheme, providing a decision basis for the collaborative scheduling module. At the same time, it responds to the feedback data from the execution and feedback modules, dynamically adjusts the weights and constraint parameters of the comprehensive objective, and realizes the closed-loop iteration of the optimization scheme.

[0019] It should be further explained that the multi-objective optimization module consists of: The optimized model building unit is responsible for building four sub-objective functions, a unified constraint library, an AHP judgment matrix, and a dynamic fuzzy membership correction model. It supports the dynamic adjustment and calculation of all parameters and weights, ensuring that the coupled logic is traceable and implementable. Weight Calculation and Correction Unit: The core is responsible for calculating the basic weights of AHP, verifying consistency, and correcting the dynamic fuzzy membership in real time, outputting the final coupling weights (updated every 15 minutes, synchronized with the scheduling cycle), which completely solves the problem of fuzzy and unreasonable simple weighted weights; Adaptive solution algorithm unit: Built-in improved NSGA-Ⅲ and adaptive particle swarm fusion algorithm, responsible for solving the comprehensive optimization objective. The single optimization solution time is ≤3s (determined according to the real-time scheduling requirements of the system, and redundancy is reserved in combination with the actual performance of the algorithm). The deviation between the optimization result and the theoretical optimal solution is ≤5% (balancing accuracy and solution efficiency, meeting the power system scheduling accuracy requirements). Uncertainty handling unit: Based on the robust optimization concept, it handles the uncertainties of new energy output, load and spot electricity price, and introduces interval parameters (the prediction error interval is 20% of the prediction value, which is determined based on the current level of wind and solar prediction technology) to avoid optimization failure caused by prediction errors. Optimization result verification unit: Performs feasibility verification on the obtained optimization scheme. After confirming that all constraints are met, the optimization scheme is output to the collaborative scheduling module in real time. If not, it returns to the algorithm unit to readjust parameters and iterate to solve. The verification cycle is consistent with the scheduling cycle (15min).

[0020] It should be further explained that the specific workflow of the multi-objective optimization module is as follows: S1: Model Initialization: Based on the standardized data from the data acquisition module, the optimization model building unit initializes the sub-objective function, unified constraint set, AHP judgment matrix, dynamic fuzzy membership correction model, and determines the scheduling cycle. Time interval ; S2: Sub-objective function construction: S201: Economic cost objective (minimization, denoted as...) ):

[0021] in, The formula for calculating the sum of local power supply costs and cross-entity power exchange costs in a microgrid is:

[0022] The total power supply (kW) of the i-th microgrid at time t is physically defined as the total electrical power supplied by the i-th microgrid to the local load at time t. The local power supply unit price of the microgrid at time t (yuan / kWh) is determined by the sales price of electricity published by the local power grid company; The mutual assistance power (kW) transmitted from the i-th microgrid to the j-th microgrid at time t is physically represented as the power allocation between microgrids. A positive value indicates that i sends power to j, and a negative value indicates that j sends power to i. The unit price for power exchange (yuan / kWh) is calculated based on line transmission loss (calculated according to line length and material, usually 0.03-0.05 yuan / kWh) + operation and maintenance cost (0.02 yuan / kWh), and is taken as 1.2 times the local power supply unit price; The energy storage charging and discharging cost (in yuan) at time t is calculated using the following formula:

[0023] The charging and discharging power (kW) of the i-th microgrid energy storage at time t is physically represented as the real-time power output / input of the energy storage device. A positive value indicates discharging and a negative value indicates charging. The unit cost of energy storage charging and discharging (yuan / kWh) is calculated based on the lithium battery charging and discharging efficiency (0.9), lifespan (1500 cycles), and purchase cost (1.2 yuan / Wh), and is taken as 0.4 yuan / kWh; The cost of standby capacity is calculated as follows:

[0024] The reserve capacity (kW) of the i-th microgrid at time t is, in physical terms, the redundant power generation capacity reserved by the microgrid to cope with load fluctuations / equipment failures. The total reserve capacity (kW) reserved at the virtual power plant level at time t, in physical terms, is the cross-microgrid reserve capacity coordinated by the virtual power plant. The unit cost of standby capacity (RMB / kW) is 0.15 RMB / kW, referencing the ancillary service price in the electricity market. The aggregated revenue of the virtual power plant at time t (in yuan) is, in physical terms, the total revenue obtained by the virtual power plant from participating in electricity market transactions and providing ancillary services; the calculation formula is as follows:

[0025] The interaction power (kW) between the virtual power plant and the main power grid at time t has the physical meaning of the power of the virtual power plant to the grid / to the grid. A positive value indicates that the power is sent to the grid (connected to the grid), and a negative value indicates that the power is purchased from the grid (disconnected from the grid). The spot electricity price at time t (yuan / kWh) is, in physical terms, the trading price of a unit of electricity in the electricity market, and is published in real time by the electricity trading center (e.g., 1.2 yuan / kWh during peak hours and 0.3 yuan / kWh during off-peak hours). The ancillary service capacity (kW) provided by the virtual power plant at time t, in physical terms, is the capacity of the virtual power plant to provide ancillary services such as frequency regulation and peak shaving to the power grid. The unit price of ancillary services (yuan / kW) is, in physical terms, the revenue generated from providing 1kW of ancillary service capacity. Based on the market price of grid ancillary services, it is set at 0.2 yuan / kW. t: Scheduling time period number (no unit), physically meaning the t-th 15-minute time period within a 24-hour scheduling cycle (t=1-96); T: Total scheduling period (h), which physically represents the time span of one optimized scheduling operation. Here, it is taken as 24h.

[0026] S202: Carbon emission target (minimization, denoted as...) ):

[0027] in: The fossil fuel power generation at time t (kWh) has the physical meaning of the power generation of fossil fuel generators such as coal and gas in the microgrid during the time period t, and is obtained by multiplying the actual output by the time interval Δt (0.25h). The carbon emission factor per unit fossil energy (kg / kWh) is, in physical terms, the CO2 emissions generated per kWh of fossil energy electricity generated. For coal-fired power generation, it is taken as 0.98 kg / kWh, and for natural gas power generation, it is taken as 0.42 kg / kWh. Total carbon emissions (kg) during the scheduling cycle, in physical terms, is the total amount of CO2 generated by the system through fossil fuel power generation within 24 hours.

[0028] S203: Power supply reliability target (maximization, denoted as...) Transform into a minimization objective for coupling purposes).

[0029] in: The total electrical load of the system at time t (kW) is physically the total electrical power consumption of all users within the coverage area of ​​the virtual power plant during time period t, and is collected by the data acquisition module. The power outage load (kW) at time t is physically defined as the load power that could not be met during time period t, which is obtained from the difference between the total load and the actual power supply. The power supply deficit rate (dimensionless) is the physical meaning of the proportion of the total load that the system fails to meet. The value ranges from [0, 1]. The smaller the value, the higher the power supply reliability.

[0030] S204: New energy absorption rate target (maximization, denoted as f4, transformed into a minimization target for coupling): in: The actual renewable energy consumption at time t (kWh) has the physical meaning of renewable energy power generation absorbed by local load / energy storage and not abandoned during time period t. It is obtained by renewable energy actual output × Δt - wind and solar curtailment. The theoretical renewable energy generation at time t (kWh) is physically the maximum electrical energy that renewable energy equipment can generate within time period t based on wind and solar resource conditions (irradiance, wind speed). It is calculated by collecting irradiance and wind speed data from the data acquisition module, combined with the rated power and power generation efficiency of the wind and solar generators. f4: Renewable energy curtailment rate, which physically represents the proportion of unconsumed renewable energy to the theoretically achievable power generation. The value ranges from [0, 1]. The smaller the value, the better the renewable energy consumption effect (e.g., f4=0 means 100% consumption). S3: Sub-target Normalization: Normalize the four sub-targets separately to the [0, 1] interval to ensure the comparability of target weights during coupling. The normalization formula is unified as follows: in: The normalized result of the i-th sub-objective is physically the sub-objective value after eliminating dimensions, and is used for multi-objective coupled calculation. The original calculated value of the i-th sub-target; The historical statistical minimum value of the i-th sub-target is, in physical terms, the minimum value that the target has taken in the past three months of operation. The historical statistical maximum value of the i-th sub-target is, in physical terms, the maximum value that the target has taken in the past three months of operation. Updated every 7 days.

[0031] S4: Strongly Coupled Core: AHP + Dynamic Fuzzy Membership Method S401 Step 1: AHP Calculation of Basic Weights: Constructing the AHP Judgment Matrix: Combining the core requirements of collaborative dispatching of virtual power plants and microgrids, and referring to power system dispatching industry standards, a 4×4 judgment matrix is ​​constructed (using the 1-9 scale method, where the scale means: 1 = both objectives are equally important, 3 = the former is slightly more important than the latter, 5 = the former is more important than the latter, 7 = the former is more important than the latter, 9 = the former is extremely important than the latter, and the reciprocal represents the opposite importance). The judgment matrix is ​​as follows: The rows and columns correspond to economic (1), low carbon (2), reliable power supply (3), and new energy consumption (4), respectively. Calculate the basic weights: Calculate the eigenvectors of the judgment matrix using the sum-product method, and then normalize them to obtain the basic weights. The calculation results are as follows: =0.5396 (economic) =0.2697 (low carbon) =0.1349 (reliable) =0.0558 (consumption); ω0: AHP basic weight vector, physically representing the static priority weights of each sub-target, satisfying...

[0032] Consistency check: Calculate the consistency index To determine the largest eigenvalue of a matrix (n=4), the following calculation is performed: =4.1045, CI=0.0348; The average random consistency index RI is 0.90 (n=4), indicating a consistency ratio of... Meets consistency requirements; Step 2 of S402: Dynamic Fuzzy Membership Correction: Based on real-time operating conditions (grid margin, renewable energy output fluctuations, load gap, carbon price fluctuations), a dynamic fuzzy membership function is constructed to correct the basic weights of AHP, resulting in the final coupling weights. The correction logic is as follows: Define fuzzy evaluation indicators: Select four real-time evaluation indicators, each corresponding to one of the four sub-objectives. The indicator values ​​are calculated from the standardized data of the data acquisition module and are all normalized to [0, 1]: Indicator 1 The economic evaluation indicator, in physical terms, is the proportion of grid interaction power to the maximum interaction power. The larger the value, the higher the priority of the economic target should be. Indicator 2 The low-carbon evaluation index, in physical terms, is the renewable energy curtailment rate. The higher the value, the higher the priority of the low-carbon target should be. Indicator 3 Safety evaluation indicators, in physical terms, represent power grid margin. The smaller the value, the closer the line is to overload, and the higher the priority of safety objectives should be. Indicator 4 The evaluation index for renewable energy consumption is, in physical terms, the proportion of renewable energy output to total load. A higher value indicates a higher priority for renewable energy consumption targets. in, This represents the maximum power exchange capacity of the power grid (kW). This represents the real-time transmission power of the line (kW). This represents the maximum transmission power of the line (kW). Total power output of new energy sources (kW); Constructing fuzzy membership functions: fuzzy evaluation indicators corresponding to the four sub-objectives of economy, low carbon, power supply reliability, and renewable energy consumption. A unified membership function for triangles is constructed, with the following general form:

[0033] The physical meaning and value rules of each parameter are as follows: The fuzzy membership degree of the i-th sub-target is physically represented as the priority correction coefficient of the target based on the corresponding real-time evaluation index, and its value range is [0, 1]. The real-time fuzzy evaluation index (dimensionless) corresponding to the i-th sub-objective; The upper and lower limits of the index range with a membership degree of 1 are physically defined as the threshold range within which the evaluation index reaches the "highest priority of the target". They are determined based on power system operation safety specifications, equipment parameters and historical operation data statistics. The critical value at which the membership degree drops to 0 is, in physical terms, the threshold at which the evaluation indicator reaches the "lowest priority of the target." Its value is determined by combining the indicator's safety boundary and operational experience. Overall rule: When the evaluation indicator is at... When the evaluation index is in a range, the corresponding sub-objective has the highest priority (membership = 1); when the evaluation index is in a range... When the evaluation index is within a certain range, the priority decreases linearly with the change of the indicator; when the evaluation index exceeds a certain range... When dealing with intervals, the priority is lowest (membership degree = 0).

[0034] Weight adjustment calculation: Final coupling weight = base weight × corresponding membership degree / sum of membership degrees, the formula is:

[0035] The final coupling weight (dimensionless) of the i-th sub-target is physically represented by the target priority ratio after combining the static base weight and the real-time running state, satisfying the following condition:

[0036] Step 3 of S403: Construct a unique comprehensive optimization objective: Multiply the four normalized sub-objectives by the final coupling weights to obtain a unique comprehensive optimization objective (minimization):

[0037] The comprehensive optimization target value, in physical terms, is the coupled optimization result of four sub-objectives: economy, low carbon, power supply reliability, and new energy consumption. The value range is [0, 1], and the smaller the value, the better the overall optimization effect of the four sub-objectives. Normalized value of economic objectives Normalized value of low-carbon target Normalized value of power supply reliability target Normalized values ​​of new energy consumption targets (all are dimensionless).

[0038] S5: Uniform Constraint Set Unified constraint set serves comprehensive objective To ensure the feasibility of the optimization plan, it is divided into three categories, each with clearly defined parameter requirements based on the actual engineering situation, as described below: Power balance constraints: Ensure system supply and demand balance, clarify the definition and basic requirements of the total output of the virtual power plant, the renewable energy output of each microgrid, and the power allocation between the virtual power plant and each microgrid at time t. The total output is determined by the sum of the output of each microgrid and the power interaction with the grid.

[0039] Equipment operating constraints (to ensure safety): New energy output constraints: The output of new energy equipment must be between the rated maximum output and the minimum technical output (the minimum output of photovoltaic is 0, and the minimum output of wind power is 12% of the rated output). Energy storage constraints: The state of charge of the energy storage should be between 0.2 and 0.8, and the charging and discharging power should be within the minimum and maximum range allowed by the equipment. The calculation logic of the state of charge and the meaning of related parameters should be clearly defined. Line constraints: Line transmission power, node voltage, and system frequency must be within their respective allowable ranges (10kV grid voltage 9.3-10.7kV, system frequency 49.8-50.2Hz).

[0040] Coordinated scheduling constraints (ensuring coordination): Power exchange constraint: The power exchange between microgrids shall not exceed the maximum allowable value of the line, and only positive exchange is allowed; Backup sharing constraint: The total backup capacity of each microgrid and virtual power plant must meet the system's backup requirements (take 8% of the total load).

[0041] S6: Dynamic optimization solution and result output: (1) The adaptive solution algorithm unit starts the improved NSGA-Ⅲ and adaptive particle swarm fusion algorithm, in order to With the objective of iteratively optimizing within a unified constraint set, the single solution time is ≤3s and the optimization accuracy is ≤5%. (2) The uncertainty processing unit introduces a prediction error range (taken as 20% of the predicted value) to robustly correct the predicted values ​​of new energy output, load and spot electricity price, so as to avoid optimization failure caused by prediction error; (3) The optimization result verification unit performs feasibility verification on the obtained scheduling scheme (power allocation, energy storage charging and discharging, and reserve allocation). After confirming that all constraints are met and consistent with the comprehensive objective, it outputs the results to the collaborative scheduling module in real time. If not, it returns to the algorithm unit to readjust the parameters and iterate to solve the problem.

[0042] The collaborative scheduling module analyzes the comprehensive optimal solution output by the multi-objective optimization module, transforms it into executable scheduling instructions, constructs a two-way interactive collaborative mechanism between the virtual power plant and the microgrid, monitors the execution effect and constraint satisfaction of instructions in real time, dynamically adjusts scheduling instructions, handles abnormal scheduling scenarios, ensures the accurate issuance and efficient execution of scheduling instructions, promotes the implementation of comprehensive optimization goals, and feeds back the execution status to the multi-objective optimization module to support the iterative optimization of comprehensive goals.

[0043] It should be further explained that the collaborative scheduling module consists of: Instruction parsing and generation unit: responsible for parsing the comprehensive optimal solution output by the multi-objective optimization module, and breaking it down into specific dispatch instructions (power allocation, energy storage charging and discharging, and reserve capacity adjustment instructions) for the virtual power plant control center and each microgrid. The instruction format adopts the power system standard protocol (IEC61850) to ensure that each execution terminal can recognize it and that the instruction parameters are fully matched with the comprehensive objectives and equipment constraints. Two-way interaction unit: Establishes a dual-redundant two-way communication channel (5G + fiber optic) between the virtual power plant and each microgrid, enabling the virtual power plant to issue dispatch commands to the microgrid, while simultaneously receiving local real-time operating status feedback from the microgrid (energy storage SOC, actual output of new energy sources, local load changes, equipment operating status) and dispatch constraints (such as energy storage failures). The communication success rate is ≥99.9%, and the command issuance delay is ≤80ms (determined based on the overall real-time requirements of the system to ensure smooth integration with the optimization and execution modules). The Cooperative Constraint Execution Unit is responsible for executing four major cooperative constraints: power mutual assistance, energy storage linkage, backup sharing, and security complementarity. It monitors the constraint fulfillment status in real time during the execution of dispatch instructions (focusing on verifying whether line transmission power, energy storage SOC, and backup capacity meet the constraints). If constraint exceedances occur, the unit will act according to the overall objective. The weight gradient adjustment instructions ensure that the overall objective is not deviated from, and the scheduling error of power mutual assistance and reserve sharing is ≤3% (determined according to the accuracy requirements of coordinated scheduling). Anomaly Dispatch Processing Unit: Monitors system operation status in real time, detects anomalies (microgrid faults, sudden drop in renewable energy output, grid overload, voltage / frequency exceeding limits), triggers tiered emergency coordination strategies, with anomaly detection delay ≤500ms, emergency strategy triggering time ≤1s, and emergency command issuance time ≤5ms (based on power system fault handling requirements to avoid fault escalation); Dispatch Strategy Adaptation Unit: Dynamically adjusts the coordinated dispatch logic based on the changes in the comprehensive objective coupling weight ω fed back by the multi-objective optimization module, ensuring that the coordinated dispatch is consistent with the priority of the comprehensive objective.

[0044] It should be further explained that the specific workflow of the collaborative scheduling module is as follows: Command parsing and generation: The command parsing and generation unit receives the comprehensive optimal solution output by the multi-objective optimization module, decomposes it into specific executable scheduling commands, and clarifies the power reception / transmission quota, energy storage charging and discharging power, and reserve capacity requirements of each microgrid, as well as the global coordination commands for the virtual power plant; among them, the power allocation command strictly follows the power balance constraints of the multi-objective optimization module, and the formula is: Where n represents the number of microgrids under the virtual power plant. Let be the total power supply of the i-th microgrid at time t. Let t be the power exchange between the virtual power plant and the main power grid. Let t be the total electrical load of the system at time t. Let t be the charging and discharging power of the i-th microgrid energy storage; the energy storage charging and discharging commands strictly follow the energy storage operation constraints. in This represents the minimum charging and discharging power for energy storage; a negative value indicates the minimum charging power, and a positive value indicates the minimum discharging power. (This refers to the maximum charge / discharge power of energy storage; negative values ​​represent maximum charging power, and positive values ​​represent maximum discharging power.) After the command is generated, it undergoes format standardization to ensure that the executing terminal can recognize it. Simultaneously, the consistency between the command parameters and the overall target is verified to prevent deviations from the intended purpose.

[0045] Command issuance and interaction: The bidirectional interaction unit issues standardized dispatch commands to the virtual power plant control center and the execution terminals of each microgrid in real time through dual redundant communication channels (5G + fiber optic), with a command issuance delay of ≤80ms; at the same time, it initiates a bidirectional interaction mechanism to receive local real-time operating status and scheduling constraints from each microgrid, and the feedback data is synchronized to the multi-objective optimization module in real time for dynamic correction of the comprehensive objective coupling weight and iteration of the optimization scheme.

[0046] Cooperative constraint monitoring: The cooperative constraint execution unit monitors the execution process of scheduling commands in real time, verifies the satisfaction of cooperative constraints such as power mutual assistance and energy storage linkage, and the command execution verification cycle is ≤500ms (meaning the command execution effect is verified every 500ms to ensure timely detection of deviations); if line overload occurs... in, Let t be the real-time transmission power of the line. The maximum transmission power of the line, determined by factors such as line material, cross-sectional area, and operating environment, represents the critical value for safe line operation. (The energy storage SOC exceeds a reasonable range.) in The real-time state of charge (SOC) of the i-th microgrid energy storage battery pack at time t is the ratio of remaining energy storage capacity to rated capacity. 0.2 and 0.8 represent the minimum and maximum SOC thresholds for safe operation of energy storage, respectively, to avoid issues such as overcharging / discharging damaging the batteries and insufficient reserve capacity. Immediately adjust the SOC according to the comprehensive target weight gradient instruction, using the following formula (taking power mutual assistance adjustment as an example): in, 0.05 is the redundancy adjustment amount, ensuring that the line transmission power is below the maximum threshold after adjustment, reserving a safety margin. To adjust the mutual power supplied by the i-th microgrid to the j-th microgrid, To adjust the mutual support power; during the adjustment process, priority will be given to targets with higher weights to ensure that the adjusted instructions still revolve around the overall objective. optimization.

[0047] Anomaly Handling: The anomaly dispatching unit monitors the system's operating status in real time and uses a threshold method to detect anomalies (frequency fluctuation > ±0.2Hz, voltage fluctuation > ±7%, sudden drop in renewable energy output > 20%, equipment failure). The frequency fluctuation threshold (±0.2Hz) and voltage fluctuation threshold (±7%) are based on power system safety operation standards. The sudden drop in renewable energy output threshold (>20%) is determined based on the characteristics of renewable energy output fluctuations. Equipment failure refers to situations where microgrids, energy storage, and transmission equipment cannot operate normally. Upon detecting an anomaly, an emergency coordination strategy is triggered within 1 second, and an emergency command is issued within 5ms. In an emergency, the coordination dispatching strategy switches to "safety priority," mobilizing the virtual power plant's reserve capacity and power mutual assistance from adjacent microgrids to quickly compensate for power shortages. Emergency reserve calls must meet the following requirements: in, The power deficit is due to a fault. (This represents the actual total output of the system at time t). For a microgrid to be operating normally, The total reserve capacity reserved at the virtual power plant level at time t. Reserved capacity for the i-th microgrid at time t; after the anomaly is resolved, automatically switch back to the conventional coordinated scheduling strategy to ensure alignment with the overall objectives. Maintain consistency.

[0048] Strategy adaptation: The scheduling strategy adaptation unit dynamically adjusts the collaborative scheduling logic based on the changes in coupling weight ω fed back by the multi-objective optimization module.

[0049] The execution and feedback module receives scheduling instructions from the collaborative scheduling module, drives the virtual power plant and related equipment in each microgrid to execute the instructions accurately, collects equipment execution status and system operation data in real time, verifies the execution effect of instructions, and transmits feedback data to the data acquisition module, multi-objective optimization module, and collaborative scheduling module to form a closed-loop feedback, supporting the dynamic iteration of comprehensive objectives and optimization schemes, and ensuring the stable operation of the entire system and the achievement of comprehensive optimization objectives.

[0050] It should be further explained that the execution and feedback module consists of the following components: Execution terminal units: Deployed in the virtual power plant control center and various microgrid nodes, including new energy power generation control terminals (wind and solar inverter control, output regulation devices), energy storage control terminals (charge and discharge controllers, SOC monitoring devices), load regulation terminals (adjustable load switches, load distribution devices), and transmission equipment control terminals (line switches, power regulation devices); the execution terminal failure rate is ≤0.1% / day (referring to the probability of execution terminal failure per day, determined based on existing power control terminal reliability indicators), and has fault self-diagnosis and self-recovery functions (fault self-diagnosis means that the terminal can automatically detect its own operational faults, and self-recovery means that minor faults can be automatically repaired without manual intervention), ensuring long-term stable operation of the equipment; Status acquisition unit: Real-time acquisition of the operating status of each execution terminal (whether the command is executed normally, equipment operating parameters) and real-time system operating data (actual power output, voltage frequency, energy storage SOC, actual consumption of new energy, line transmission power). The acquisition frequency is consistent with that of the data acquisition module (new energy output and load data 1 min / time, power grid status data 500 ms / time, market data 15 min / time) to ensure data synchronization. Feedback transmission unit: Establishes encrypted communication with the data acquisition module, multi-objective optimization module, and collaborative scheduling module (using AES-256 encryption to prevent data leakage or tampering), and feeds back the real-time execution status, system operation data, and execution verification results to the corresponding modules. The feedback delay is ≤50ms, ensuring that the feedback data supports the adjustment of the comprehensive objectives and optimization schemes in a timely manner. Execution verification unit: Verifies the execution effect of scheduling instructions, calculates execution deviation, determines whether there is execution anomaly, and immediately triggers anomaly feedback when an anomaly occurs. The execution verification cycle is ≤500ms to ensure that execution deviation can be detected and handled in a timely manner.

[0051] It should be further explained that the specific workflow of the execution and feedback module is as follows: Command Reception: The execution terminal unit receives standardized dispatch commands issued by the collaborative dispatch module (the command format adopts the power system standard protocol IEC61850 to ensure recognizability), parses the command content (energy storage charging and discharging power, new energy output adjustment value, load allocation ratio, etc.), verifies the command format and parameter range, and confirms that the command parameters are consistent with equipment operating constraints (such as energy storage charging and discharging power not exceeding the rated range) and comprehensive objectives. (The overall optimization target value is dimensionless and ranges from [0, 1]. The smaller the value, the better the overall optimization effect of the four sub-targets.) Once the target value is consistent, the equipment execution process is started. If the instruction parameters are abnormal (exceeding the constraint range), the execution is refused and the feedback is immediately sent to the collaborative scheduling module.

[0052] Instruction Execution: Each execution terminal drives its corresponding device to strictly execute the scheduling instructions. The execution process strictly follows the constraints of the multi-objective optimization module to ensure that the execution effect meets the overall objective. Consistency: New energy power generation control terminal: Adjusts the output power of wind and solar inverters according to instructions, with adjustment accuracy meeting requirements. in, To make practical contributions to new energy, The output adjustment response time is ≤200ms (the time interval from receiving the adjustment command to the actual output adjustment being completed, determined based on the inverter's response characteristics), ensuring that the output of new energy meets the requirements of consumption and low carbon in the comprehensive objectives; the energy storage control terminal controls the charging and discharging status and charging and discharging power of the energy storage equipment according to the command, maintaining the energy storage SOC within a reasonable range of [0.2, 0.8] (SOC is the state of charge of energy storage, referring to the proportion of the remaining energy of the energy storage to the rated capacity; 0.2 and 0.8 are safety thresholds to avoid overcharging and discharging damage to the battery), and the charging and discharging power adjustment response time is ≤1 ... charging and discharging command to the actual output adjustment being completed, determined based on the inverter's response characteristics), ensuring that the output of new energy meets the requirements of consumption and low carbon in the comprehensive objectives); the energy storage control terminal controls the charging and discharging status and charging and discharging power of the energy storage equipment according to the command, maintaining the energy storage SOC within a reasonable range of [0.2, 0.8] (SOC is the state of charge of energy storage, referring to the proportion of the remaining energy of the energy storage to the rated capacity; 0.2 and 0.8 are safety thresholds to avoid overcharging and discharging damage to The time interval between the discharge command and the power adjustment is determined based on the performance of the energy storage controller, ensuring that the energy storage linkage meets the power balance and absorption targets; the load control terminal adjusts the power consumption period and power of the adjustable load according to the command, with a load adjustment error ≤2% (referring to the ratio of the deviation between the actual load adjustment value and the command requirement value to the command value, determined based on the performance of the load control equipment), balancing local load and output, improving power supply reliability, and meeting the safety requirements in the comprehensive target; the transmission equipment control terminal adjusts the line transmission power according to the command, ensuring reasonable power mutual assistance, avoiding line overload, and the adjustment accuracy meets the collaborative dispatch error ≤3% (consistent with the accuracy requirements of the collaborative dispatch module), meeting the safety and economic requirements in the comprehensive target.

[0053] Status Acquisition and Verification: The status acquisition unit collects the execution status and system operation data of each device in real time and transmits them synchronously to the execution verification unit; the execution verification unit compares the instruction requirements with the actual execution effect, calculates the execution deviation, and the deviation calculation formula is: Where δ is the execution deviation, which is dimensionless and reflects the degree of deviation between the actual execution value and the instruction value; when δ>2%, it is judged as an execution anomaly, and anomaly feedback is immediately triggered. At the same time, the anomaly data (deviation value, anomaly time, equipment status) is recorded for subsequent parameter adjustment; the execution verification cycle is ≤500ms to ensure that anomalies can be detected in a timely manner and avoid affecting the achievement of the overall goal.

[0054] Feedback Transmission: The feedback transmission unit feeds back the real-time collected execution status, system operation data, and execution verification results (normal / abnormal, deviation value) to the data acquisition module (for data updates and historical storage), the multi-objective optimization module (for comprehensive objective coupling weight correction and optimization scheme iteration), and the collaborative scheduling module (for scheduling command correction), respectively; the feedback data includes actual output. Core parameters such as energy storage SOC, voltage, frequency, and execution deviation δ have a feedback delay of ≤50ms and are transmitted using encrypted transmission (AES-256 encryption) to ensure data security and integrity, providing support for closed-loop optimization.

[0055] Closed-loop optimization: The system receives the comprehensive optimization target (weight correction), optimization scheme, and scheduling instructions adjusted by the upstream module based on feedback data. It then repeats the "receive-execute-collect-verify-feedback" process to achieve a closed-loop operation of "execute-feedback-adjust-re-execute," ensuring the system always revolves around the comprehensive target. Iterative optimization to continuously improve the optimization effect.

[0056] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization, characterized in that, include: Data acquisition module: performs real-time, multi-dimensional data acquisition, anomaly cleaning, missing data completion, and dimensional unification processing of distributed renewable energy output, electricity load, energy storage operation status, grid node parameters, and electricity market and carbon price information to form standardized time-series data; Multi-objective optimization module: Based on standardized time series data, four sub-objectives are constructed: economic cost, carbon emissions, power supply reliability, and new energy consumption. Static basic weights are determined by the analytic hierarchy process and dynamic fuzzy membership is used to adaptively correct the weights in combination with the real-time operation status of the power grid. Multiple objectives are strongly coupled into a single comprehensive optimization objective. The global optimal scheduling strategy is obtained by solving under the constraints of equipment and power grid safety. The collaborative scheduling module parses the globally optimal scheduling strategy into standardized scheduling instructions, which are then sent to each microgrid and terminal device through dual redundant communication channels. It monitors line power, energy storage status, and reserve capacity constraints in real time. In cases of exceeding limits or failures, it dynamically adjusts the scheduling instructions according to the priority of comprehensive objectives and activates a safety-first emergency collaborative mechanism. Execution feedback module: used to drive new energy, energy storage, load and transmission equipment to execute scheduling commands, collect the actual operating status and command execution deviation in real time, and synchronously transmit the operating data and verification results back to the data acquisition unit and multi-objective optimization unit to form a closed-loop iterative operation system.

2. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 1, characterized in that: The multi-objective optimization module constructs an economic cost sub-objective, including: Based on standardized time-series data, the system first calculates the product of the unit energy cost of local power supply and the total power supply of each microgrid, the product of the unit transmission cost and the mutual power of power exchange between microgrids, the product of the unit energy consumption cost and the charging and discharging power of energy storage in each microgrid, and the product of the unit reserve cost and the reserve capacity of each microgrid and virtual power plant, summing them up to obtain the total system expenditure cost. Then, the system calculates the electricity market transaction revenue corresponding to the power exchange between virtual power plants and the main grid, and the service revenue corresponding to the ancillary services provided by virtual power plants, summing them up to obtain the total system revenue. Finally, by subtracting the total revenue from the total expenditure cost, the system constructs an economic cost sub-objective oriented towards minimizing economic costs.

3. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 1, characterized in that: The multi-objective optimization module constructs carbon emission sub-objectives, including: Based on standardized time-series data, the actual power generation of fossil energy power generation equipment in each microgrid is first counted within the scheduling cycle. Combined with the carbon emission coefficient per unit power generation of the corresponding fossil energy type, the carbon emissions generated by fossil energy power generation in each time period are calculated. Then, the carbon emissions of all time periods within the scheduling cycle are summarized to construct a carbon emission sub-objective guided by minimizing the total carbon emissions.

4. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 1, characterized in that: The multi-objective optimization module constructs a power supply reliability sub-objective, including: Based on standardized time-series data, the total power load of the system within the scheduling cycle is statistically analyzed, and the total power outage load that could not be met within the same scheduling cycle is also calculated. By comparing the total power load with the total power outage load, the system power supply deficit rate is obtained. Minimizing the power supply deficit rate is used as a guide to construct the power supply reliability sub-objective.

5. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 1, characterized in that: The multi-objective optimization module constructs a new energy consumption sub-objective, including: The theoretical total power generation of distributed renewable energy within the statistical scheduling cycle is calculated, and the renewable energy power actually consumed and utilized within the same cycle is also calculated. By comparing the actual power consumption with the theoretical power generation, the renewable energy curtailment rate of the system is obtained. Taking the minimization of the renewable energy curtailment rate as the guide, a renewable energy consumption sub-objective is constructed.

6. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 1, characterized in that: The determination of static basic weights using the analytic hierarchy process includes: First, based on the requirements of coordinated dispatch of virtual power plants and microgrids and industry standards, a judgment matrix is ​​constructed for four sub-objectives: economic cost, carbon emissions, power supply reliability, and renewable energy consumption. The scaling method is used to clarify the relative importance of each sub-objective. Then, the judgment matrix is ​​solved by the eigenvector calculation method, and the obtained eigenvectors are normalized to obtain the initial weights corresponding to each sub-objective. Finally, the rationality of the judgment matrix and initial weights is verified by consistency check. The initial weights that pass the check are the static basic weights of each sub-objective.

7. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 1, characterized in that: The method of using dynamic fuzzy membership degrees for adaptive weight correction includes: First, real-time evaluation indicators corresponding to the four sub-objectives are selected. These real-time evaluation indicators are all derived from standardized time-series data and are normalized. Then, a triangular fuzzy membership function is constructed for each real-time evaluation indicator. The fuzzy membership degree corresponding to each sub-objective is calculated through this function. The fuzzy membership degree is used to represent the real-time priority of each sub-objective in the current system operating state. Finally, the static basic weight of each sub-objective is multiplied by the corresponding fuzzy membership degree, and then normalized to obtain the corrected final coupling weight.

8. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 7, characterized in that: The globally optimal scheduling strategy includes: Guided by a single comprehensive optimization objective, and under the premise of meeting the constraints of equipment operation safety, power grid transmission safety and collaborative scheduling, combined with the corrected final coupling weight, the power supply allocation of each microgrid, the power mutual assistance quota between microgrids, the charging and discharging status and power of each energy storage device, the reserve capacity allocation between each microgrid and the virtual power plant, and the interactive power between the virtual power plant and the main power grid are planned to form a complete scheduling scheme.

9. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 1, characterized in that: The emergency coordination mechanism includes: When a grid operation exceeds limits, equipment failure, or fluctuations in new energy output are detected, the system switches to a safety-priority dispatch mode, calls upon the reserve capacity of virtual power plants and normally operating microgrids, adjusts the power mutual assistance relationship between microgrids, and simultaneously corrects the current dispatch instructions. This collaborative approach restores stable system operation while ensuring the safety of the grid and equipment.

10. The virtual power plant and microgrid collaborative dispatch system based on multi-objective optimization according to claim 1, characterized in that: The closed-loop iterative operation system includes: The actual running status returned by the execution feedback module and the deviation from the instruction execution are synchronously connected to the data acquisition module for standardization processing, and then input to the multi-objective optimization module to update the target weights and optimize the scheduling strategy. The correction instruction is issued through the collaborative scheduling module, and then executed and fed back by the execution feedback module, forming a cyclical iterative running mode of data acquisition, optimization calculation, scheduling issuance, and execution verification.

Images