Micro-grid group operation optimization method and system based on multi-target dragonfly algorithm

Abstract

The invention discloses a micro-grid group operation optimization method and system based on a multi-target dragonfly algorithm, and the method comprises the steps: constructing a system model of a micro-grid cluster, and enabling the system model to comprise a plurality of micro-grids; taking the connection state of the plurality of micro-grids as a decision variable to represent the communication mode between the micro-grids; a dynamic grouping mechanism is introduced, and the connection relation of the multiple micro-grids is adjusted according to actual requirements; establishing a multi-objective optimization model, and designing operation constraint conditions; combining decision variables and constraint conditions, designing a structure-scheduling joint optimization strategy, and carrying out collaborative optimization on a micro-grid group structure and a scheduling strategy; and according to an optimization result, outputting a Pareto optimization scheduling solution set covering different operation preferences, and obtaining a low-carbon economic multi-energy-flow cooperative operation scheme of the micro-grid group. According to the method, the problems of local optimum and low convergence speed during multi-objective optimization in the micro-grid dispatching process can be solved.

Description

Technical Field

[0001] This invention belongs to the field of microgrid group optimization operation technology, and particularly relates to a microgrid group operation optimization method and system based on multi-objective dragonfly algorithm. Background Technology

[0002] With the rapid development of renewable energy, microgrids, as an important distributed energy management system, have demonstrated enormous potential in achieving efficient energy utilization, reducing carbon emissions, and improving the stability of the power system. However, during the operation of microgrid clusters, the increasing penetration rate of renewable energy (such as photovoltaic and wind power) brings increasing challenges to the system. Traditional microgrid dispatching methods often rely on fixed network structures and single optimization objectives, making it difficult to cope with variable load fluctuations and the uncertainties of distributed energy sources. This uncertainty directly affects the resource dispatching and energy interaction strategies of microgrids, thereby impacting the system's operational efficiency and sustainability. On the one hand, as the scale of microgrid clusters expands, the connection and disconnection between microgrids need to be flexibly adjusted according to real-time load demand and changes in renewable energy generation. On the other hand, microgrid dispatching not only needs to consider economic efficiency but also needs to comprehensively consider multiple objectives such as carbon emissions, environmental impact, and renewable energy consumption, transforming the dispatching problem from a single-objective optimization problem into a multi-objective optimization problem.

[0003] In the scheduling optimization of microgrid clusters, existing research often employs enumeration methods to combine all possible microgrid cluster configurations. While this method comprehensively considers microgrid scheduling strategies, its high computational complexity often limits its application to large-scale microgrid clusters with complex scheduling strategies. In designing the objective function, most studies focus on optimizing operating and environmental costs, such as fuel costs, electricity purchase costs, and carbon emission costs. However, with the increasing penetration of renewable energy, the curtailment costs in microgrid cluster scheduling (e.g., the cost of curtailing solar and wind power) are becoming increasingly significant. Incorporating renewable energy curtailment costs into the objective function aims to maximize renewable energy utilization and reduce curtailment while lowering operating costs. Regarding optimization methods, existing approaches often rely on traditional heuristic algorithms, such as genetic algorithms and particle swarm optimization, to address the complexity of microgrid cluster scheduling. However, these methods typically suffer from local optima and slow convergence speeds, especially when facing multi-objective optimization problems, often failing to guarantee a globally optimal solution. Summary of the Invention

[0004] To address the shortcomings of existing scheduling methods in flexible scheduling and multi-objective optimization of microgrid clusters, and to optimize system economy, low carbon emissions, and renewable energy integration, this invention provides a microgrid cluster operation optimization method and system based on a multi-objective dragonfly algorithm. By introducing a dynamic clustering mechanism, the microgrid can flexibly adjust its connection relationships under different operating conditions, improving the system's adaptability and flexibility. A multi-objective scheduling optimization model is proposed, comprehensively considering operating costs, carbon emission costs, and renewable energy curtailment costs, optimizing the trade-offs among multiple objectives. An improved multi-objective dragonfly algorithm is adopted, combined with chaotic mapping and nonlinear inertial weighting strategies, to enhance the algorithm's global search capability and local convergence accuracy. By adjusting the microgrid cluster networking and scheduling scheme through operating data and scheduling feedback, the system can achieve optimal scheduling performance under different operating conditions.

[0005] The present invention adopts the following technical solution.

[0006] This invention provides a microgrid group operation optimization method based on the multi-objective dragonfly algorithm, comprising: Step 1: Construct a system model of a microgrid cluster, which includes multiple microgrids; use the connection status of the multiple microgrids as a decision variable to represent the interconnection mode between the microgrids; introduce a dynamic grouping mechanism to adjust the connection relationship of the multiple microgrids according to actual needs; Step 2: Establish a multi-objective optimization model that comprehensively considers minimizing operating costs, carbon emissions, and renewable energy curtailment losses, and design operating constraints including power balance, energy storage limitations, and power exchange capacity. Step 3: Combining the decision variables and constraints, design a structure-scheduling joint optimization strategy to collaboratively optimize the microgrid group structure and scheduling strategy; Step four: Based on the optimization results, output the Pareto optimized scheduling solution set covering different operating preferences to obtain the multi-energy flow coordinated operation scheme of the microgrid group for low-carbon economy.

[0007] More preferably, in step one, binary variables are used. To indicate whether microgrids are connected, i.e. ;when Represents the microgrid at time t and microgrids There are connections between them; Represents the microgrid at time t and microgrids No connection.

[0008] More preferably, in step one, the dynamic grouping mechanism allows the microgrid to adaptively adjust its networking mode within the scheduling cycle or according to changes in system load; The dynamic grouping mechanism is specifically expressed as follows: under the premise of satisfying the transmission capacity of the connection lines, the microgrid... and microgrids The connection status adjustment process is as follows: ; ; , ; in, microgrid and microgrids The complementary needs between them serve as the basis for determining the connection; Indicates the connection hysteresis compensation threshold; Represents a binary variable, indicating whether microgrids are connected; This represents a binary variable adjusted through a dynamic grouping mechanism. This indicates the line loss caused by switching power.

[0009] More preferably, in step one, when the microgrid and microgrids Interconnection, under the constraint of meeting the transmission capacity of the connection lines, microgrids and The calculation of power exchange between them is expressed by the following equation: ; in: microgrid and The actual amount of power exchanged between them; Represents a binary variable, indicating whether microgrids are connected; and They represent microgrids and The output power; and They represent microgrids and The electrical load; Indicates the maximum power transmission capacity of the connecting line; This indicates the line loss caused by switching power.

[0010] More preferably, in step three, the structure-scheduling joint optimization strategy adopts an improved multi-objective dragonfly algorithm, which uses structural variables and scheduling variables to jointly optimize the multi-objective optimization model of the microgrid cluster by simulating the behavior rules of the dragonfly swarm; The decision variables are used as structural variables, and the constraint-based scheduling strategy is used as the scheduling variable.

[0011] More preferably, in step three, the improved multi-objective dragonfly algorithm is improved by using chaotic mapping and nonlinear inertial weighting strategies; the dragonfly population is initialized by chaotic mapping, and the update step size of individual dragonflies is adjusted by nonlinear inertial weighting to update the position of the dragonflies.

[0012] More preferably, in step three, the updating of the dragonfly's position is related to the dragonfly's separation behavior, alignment behavior, aggregation behavior, foraging behavior, and predator avoidance behavior. Based on the heterogeneity of structural variables and scheduling variables, the control of structural variables and scheduling variables is performed in a domain-specific manner.

[0013] More preferably, in step three, the structural variables and scheduling variables are packaged into a single individual code, which is then used as the object of optimization by the Dragonfly algorithm; In the optimization process, domain-specific processing refers to adjusting structural variables using separation and aggregation behaviors, and optimizing operational variables using alignment, foraging, and enemy avoidance behaviors. The results of the domain optimization are merged to form the overall individual changes, and the position of the entire dragonfly is updated uniformly.

[0014] This invention also proposes a microgrid cluster operation optimization system based on the multi-objective dragonfly algorithm using the above method, including a microgrid cluster system modeling module, a multi-objective optimization modeling module, a structure-scheduling joint optimization module, and an optimization solution output module: The microgrid cluster system modeling module constructs a system model of a microgrid cluster, which includes multiple microgrids; the connection status of the multiple microgrids is used as a decision variable to characterize the interconnection mode between the microgrids. The multi-objective optimization modeling module establishes a multi-objective optimization model that comprehensively considers minimizing operating costs, carbon emissions, and renewable energy curtailment losses, and designs operating constraints including power balance, energy storage limitations, and power exchange capacity. The structure-scheduling joint optimization module combines decision variables and constraints, and adopts an improved multi-objective dragonfly algorithm to design a structure-scheduling joint optimization strategy, thereby achieving coordinated optimization of group structure and scheduling strategy and search for the optimal solution set. The optimized solution set output module outputs Pareto optimized scheduling solutions covering different operating preferences, resulting in a multi-energy flow coordinated operation scheme for low-carbon economy of microgrid groups.

[0015] The present invention also proposes a terminal, including a processor and a storage medium: The storage medium is used to store instructions; The processor is used to perform the steps of the above method according to the instructions.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention effectively enhances the collaborative scheduling capability of microgrid clusters by introducing an improved multi-objective dragonfly optimization algorithm. By accurately modeling the energy structure and operational constraints of the microgrid, the microgrid can better cope with renewable energy fluctuations, flexibly adjust power flow and energy storage scheduling, and optimize energy consumption paths in advance. By introducing curtailment costs and a dynamic clustering mechanism, the renewable energy absorption rate is improved, the system's adaptability and stability are enhanced, and scheduling conflicts caused by energy fluctuations and prediction errors are reduced, thereby achieving more efficient and low-carbon microgrid cluster operation.

[0017] 2. This invention introduces a dynamic grouping mechanism, allowing microgrids to flexibly adjust their connection relationships under different operating conditions, thus improving the system's adaptability and flexibility. It proposes a multi-objective scheduling optimization model that comprehensively considers operating costs, carbon emission costs, and renewable energy curtailment costs, optimizing the trade-offs among multiple objectives. An improved multi-objective dragonfly algorithm, combined with chaotic mapping and nonlinear inertial weighting strategies, enhances the algorithm's global search capability and local convergence accuracy. By adjusting the microgrid grouping and scheduling scheme through operating data and scheduling feedback, the invention ensures that the system achieves optimal scheduling performance under different operating conditions.

[0018] 3. This invention significantly improves the coordinated scheduling performance of microgrids under complex constraints by constructing a joint structure-scheduling optimization strategy. By guiding structural and operational variables with different optimization behaviors, it solves the problem of traditional algorithms struggling to simultaneously optimize both structure and scheduling. Furthermore, by combining a dynamic swarming mechanism with the multi-objective dragonfly algorithm, it effectively addresses the coupling problem in energy scheduling, improving the stability and convergence speed of the optimization. This innovative approach not only improves scheduling accuracy but also enhances the algorithm's adaptability and efficiency, providing a new solution for the optimized scheduling of microgrids. Attached Figure Description

[0019] Figure 1 This is a flowchart of a microgrid group operation optimization method based on the multi-objective dragonfly algorithm of the present invention; Figure 2 This is a flowchart of the method according to Embodiment 1 of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0021] The present invention proposes the following technical solution.

[0022] like Figure 1 As shown, this invention proposes a microgrid group operation optimization method based on the multi-objective dragonfly algorithm, including: Step 1: Construct a system model of a microgrid cluster, which includes multiple microgrids; use the connection status of the multiple microgrids as a decision variable to represent the interconnection mode between the microgrids; introduce a dynamic grouping mechanism to adjust the connection relationship of the multiple microgrids according to actual needs; In step one, binary variables are used. To indicate whether microgrids are connected, i.e. ;when Represents the microgrid at time t and microgrids There are connections between them; Represents the microgrid at time t and microgrids No connection.

[0023] In step one, the dynamic grouping mechanism allows the microgrid to adaptively adjust its networking mode within the scheduling cycle or according to changes in system load; The dynamic grouping mechanism is specifically expressed as follows: under the premise of satisfying the transmission capacity of the connection lines, the microgrid... and microgrids The connection status adjustment process is as follows: ; ; , ; in, microgrid and microgrids The complementary needs between them serve as the basis for determining the connection; Indicates the connection hysteresis compensation threshold; Represents a binary variable, indicating whether microgrids are connected; This represents a binary variable adjusted through a dynamic grouping mechanism. This indicates the line loss caused by switching power.

[0024] In step one, when the microgrid and microgrids Interconnection, under the constraint of meeting the transmission capacity of the connection lines, microgrids and The calculation of power exchange between them is expressed by the following equation: ; in: microgrid and The actual amount of power exchanged between them; Represents a binary variable, indicating whether microgrids are connected; and They represent microgrids and The output power; and They represent microgrids and The electrical load; Indicates the maximum power transmission capacity of the connecting line; This indicates the line loss caused by switching power.

[0025] Step 2: Establish a multi-objective optimization model that comprehensively considers minimizing operating costs, carbon emissions, and renewable energy curtailment losses, and design operating constraints including power balance, energy storage limitations, and power exchange capacity. Step 3: Combining the decision variables and constraints, design a structure-scheduling joint optimization strategy to collaboratively optimize the microgrid group structure and scheduling strategy; In step three, the structure-scheduling joint optimization strategy adopts an improved multi-objective dragonfly algorithm, which uses structural variables and scheduling variables to jointly optimize the multi-objective optimization model of the microgrid cluster by simulating the behavior rules of the dragonfly swarm. The decision variables are used as structural variables, and the constraint-based scheduling strategy is used as the scheduling variable.

[0026] In step three, the improved multi-objective dragonfly algorithm is improved by using chaotic mapping and nonlinear inertial weighting strategies. The dragonfly population is initialized using chaotic mapping, and the update step size of individual dragonflies is adjusted and the position of the dragonflies is updated using nonlinear inertial weighting.

[0027] In step three, the updating of the dragonfly's position is related to the dragonfly's separation behavior, alignment behavior, gathering behavior, foraging behavior, and enemy avoidance behavior. Based on the heterogeneity of structural variables and scheduling variables, the control of structural variables and scheduling variables is performed in a domain-specific manner.

[0028] In step three, the structural variables and scheduling variables are packaged into a single individual code, which is then used as the object of optimization by the Dragonfly algorithm. In the optimization process, domain-specific processing refers to adjusting structural variables using separation and aggregation behaviors, and optimizing operational variables using alignment, foraging, and enemy avoidance behaviors. The results of the domain optimization are merged to form the overall individual changes, and the position of the entire dragonfly is updated uniformly.

[0029] Step four: Based on the optimization results, output the Pareto optimized scheduling solution set covering different operating preferences to obtain the multi-energy flow coordinated operation scheme of the microgrid group for low-carbon economy.

[0030] This invention also proposes a microgrid cluster operation optimization system based on the multi-objective dragonfly algorithm using the above method, including a microgrid cluster system modeling module, a multi-objective optimization modeling module, a structure-scheduling joint optimization module, and an optimization solution output module: The microgrid cluster system modeling module constructs a system model of a microgrid cluster, which includes multiple microgrids; the connection status of the multiple microgrids is used as a decision variable to characterize the interconnection mode between the microgrids. The multi-objective optimization modeling module establishes a multi-objective optimization model that comprehensively considers minimizing operating costs, carbon emissions, and renewable energy curtailment losses, and designs operating constraints including power balance, energy storage limitations, and power exchange capacity. The structure-scheduling joint optimization module combines decision variables and constraints, and adopts an improved multi-objective dragonfly algorithm to design a structure-scheduling joint optimization strategy, thereby achieving coordinated optimization of group structure and scheduling strategy and search for the optimal solution set. The optimized solution set output module outputs Pareto optimized scheduling solutions covering different operating preferences, resulting in a multi-energy flow coordinated operation scheme for low-carbon economy of microgrid groups.

[0031] The present invention also proposes a terminal utilizing the above method, comprising a processor and a storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps according to the method described above.

[0032] Example 1 This invention proposes a microgrid group operation optimization method based on the multi-objective dragonfly algorithm, such as... Figure 2 As shown, the specific steps are as follows.

[0033] Step 1: Construct a system model of the microgrid cluster. The system model includes multiple microgrids that are interconnected. The connection status of the multiple microgrids is used as a decision variable to represent the interconnection mode between the microgrids.

[0034] Specifically, the system model of the microgrid cluster consists of multiple microgrids interconnected through advanced communication and control technologies. The main components of the microgrid include distributed power sources, energy storage devices, energy conversion devices, and loads, which form a collaborative system. This invention uses the connection status of each part in the system as a decision variable to characterize the interconnection mode between microgrids, and introduces a dynamic networking mechanism to adapt to the diverse collaborative needs of the microgrid cluster.

[0035] Microgrid clusters can achieve optimized energy allocation and efficient utilization by coordinating the operating states of individual microgrids. These microgrids can operate independently or coordinate with each other as needed to jointly provide power supply and energy management. The architecture of microgrids involves various energy sources, which can be divided into controllable distributed energy sources and uncontrollable distributed energy sources according to the control type.

[0036] Controllable distributed energy sources mainly include micro gas turbines, and energy storage systems can also be considered a type of distributed controllable energy source. Uncontrollable distributed energy sources mainly include distributed photovoltaic power generation and decentralized wind power generation. Differential modeling is performed based on the operating characteristics of various distributed energy sources; the specific power generation mathematical model is as follows: (1) Photovoltaic power generation model (PV): ; in: This indicates the actual output power of the photovoltaic array; Indicates the first Microgrids; This represents the reference output power of the photovoltaic array under standard test conditions; Indicates the power temperature coefficient; This refers to the operating temperature of the photovoltaic array. This is the reference temperature under standard test conditions; Indicates the area of ​​the photovoltaic panel; Indicates photovoltaic capacity efficiency; Indicates the actual light intensity; This represents the reference illumination intensity under standard test conditions.

[0037] (2) Wind power generation model (WT): ; in: This indicates the actual output power of the fan; Indicates the current actual wind speed; Indicates the fan cut-in wind speed; Indicates the fan cut-out speed; This indicates the wind speed at which the fan reaches its rated power. This indicates the standard output power of the fan at its rated wind speed.

[0038] (3) Micro Gas Turbine Model (MT): ; in: Indicates the output power of the micro gas turbine; Indicates the power generation efficiency of a micro gas turbine; These represent the power generation coefficients of the micro gas turbine at different sub-terms, and are set according to actual conditions.

[0039] (4) Energy storage system model (BT): ; in: Indicates the remaining charge capacity of the battery; Indicates the charging / discharging power of energy storage; Indicates the rated capacity of energy storage; This indicates the charging / discharging efficiency of energy storage.

[0040] Furthermore, the present invention uses the connection status between the microgrids as a decision variable and introduces a dynamic grouping mechanism to enable the microgrids to adjust their connection relationship according to actual needs. This mechanism allows the microgrid group to flexibly adapt to diverse collaborative needs based on real-time load, distributed energy generation, and operating conditions.

[0041] Specifically, the connection status between the microgrids is used as a decision variable, allowing the connection relationship between microgrids to be adjusted as needed during the scheduling optimization process. A binary variable is used to represent whether the microgrids are connected, i.e.: .when Represents the microgrid at time t and microgrids There are connections between them; Represents the microgrid at time t and microgrids No connection.

[0042] The connectivity status determines whether power can be exchanged between microgrids, thus affecting the scheduling decisions of each microgrid. If the microgrid... and microgrids Interconnection, then Under the constraint of meeting the transmission capacity of the connecting lines, it is possible to implement this in a microgrid. and Energy transfer and power exchange between them can be calculated using the following equation: ; in: microgrid and The actual amount of power exchanged between them; and They represent microgrids and The output power; and They represent microgrids and The electrical load; Indicates the maximum power transmission capacity of the connecting line; Indicates the line loss caused by the exchange power.

[0043] Furthermore, the present invention designs a dynamic grouping mechanism, allowing the microgrid to adaptively adjust the networking mode within the scheduling period or according to the change of the system load.

[0044] Specifically, the microgrid decides whether to remain connected, disconnect, or reconnect according to its load demand and distributed energy generation in different time periods. If the load demand increases or the renewable energy generation fluctuates, the microgrid can compensate for the power difference by adjusting the connection state and reduce the load loss. On the premise of meeting the transmission capacity of the connection line, the microgrid and the microgrid The connection state adjustment process is as follows: ; ; , ; Among them, represents the complementary demand between microgrids i and j and is used as the judgment basis for connection; represents the connection hysteresis compensation threshold, which is the additional power complementary amount exceeding the line loss required to establish a connection in the disconnected state. If the power deficit value of microgrid i is a, the excess value of microgrid j is b, and a > b, then the complementary demand is b; if a = b, then the complementary demand is a or b; if a < b, then the complementary demand is a.

[0045] Step 2, establish a multi-objective optimization model that comprehensively considers minimizing the operating cost, carbon emissions, and renewable energy curtailment loss, and design operating constraint conditions including power balance, energy storage limit, and power exchange capacity; Specifically, construct a multi-objective optimization model including the operating cost , carbon emission cost , and renewable energy curtailment cost , and introduce constraint conditions such as power balance, energy storage operation, and power exchange capacity to achieve low-carbon economic dispatch optimization of the microgrid group. The multi-objective optimization function is as follows: ; Among them, represents the total cost objective function; represents the weight coefficients of the operating cost, carbon emission cost, and renewable energy curtailment cost.

[0046] Comprehensively consider the operating cost, carbon emission cost, and renewable energy curtailment cost, and construct a low-carbon economic optimal dispatch model for the microgrid group to minimize the total cost objective function. Among them, the operating cost includes the power generation cost of controllable distributed generation units Equipment operation and maintenance costs Electricity purchase and sale transaction costs Carbon emission costs include the penalty costs for pollutants such as carbides, sulfides, and nitrogen oxides; renewable energy curtailment costs include the costs of curtailment losses from distributed photovoltaic and decentralized wind power. Details are as follows: (1) Operating costs: ; ; ; ; in: Indicates the number of microgrids; Indicates the number of operating hours; and These represent fuel price and calorific value, respectively. These represent the equipment operation and maintenance cost coefficients for WT, PV, MT, and BT, respectively. These represent the output power of WT, PV, MT, and BT, respectively. and These represent the MG and the power purchase and sale price of the distribution network, respectively. and These represent the interaction power between MGs and between MGs and the distribution network, respectively. A positive value indicates power purchase, while a negative value indicates power sale.

[0047] (2) Carbon emission costs: ; in: Indicates the number of pollutant types; Indicates the first The cost of treating various types of pollutants; and These represent the first generation generated by MT and the second generation generated when purchasing electricity from the distribution network, respectively. Emission coefficients for pollutants of this type.

[0048] (3) Costs of abandoning renewable energy ; in: and These represent the cost coefficients for wind curtailment and solar curtailment, respectively. and These represent the predicted wind power output and the predicted solar power output, respectively. Using historical values ​​of wind and solar power output, a machine learning model is used to predict the predicted wind and solar power output. In this embodiment, the machine learning model uses a gated recurrent unit network (GRU). Since the prediction model and method are existing technologies, they will not be described in detail here.

[0049] Furthermore, a multi-objective optimization model for the microgrid cluster is constructed with minimizing the total cost as the objective function. This model encompasses constraints such as power balance, energy storage limitations, and carbon emission constraints to ensure the rational operation of the microgrid cluster. Details are as follows: (1) System power balance constraints: ; in: Indicates the first microgrids Total electrical load at any given time.

[0050] (2) Output upper and lower limits and ramping constraints of micro gas turbines: ; ; ; in: and These represent the upper and lower limits of the active power of the micro gas turbine, respectively. and These represent the maximum ramp power and maximum landslide power of the micro gas turbine, respectively.

[0051] (3) Uncontrollable distributed energy output constraints: ; ; ; ; in: and These represent the upper limits of wind power and solar power output, respectively.

[0052] (4) Constraints on energy storage state of charge and charge / discharge power ; ; ; ; in: and These represent the minimum and maximum states of charge of energy storage, respectively; and These represent the minimum and maximum charge / discharge power of the energy storage, respectively. and These represent the energy storage charging and discharging power, respectively. and These represent the state variables of energy storage charging and discharging, respectively.

[0053] (5) Interactive power constraints between MGs and between MGs and the distribution network ; ; in: microgrid and microgrids The maximum allowable interaction power between them; microgrid With distribution network The maximum allowable interaction power between them.

[0054] (6) MG Interconnection Constraints ; in: This indicates the line loss caused by power exchange. When the power exchanged between microgrids is greater than the energy loss caused by transmission lines, microgrids can be interconnected and exchange power.

[0055] Step 3: Combining decision variables and constraints, an improved multi-objective dragonfly algorithm is used to design a structure-scheduling joint optimization strategy to achieve coordinated optimization of group structure and scheduling strategy and search for the optimal solution set.

[0056] Specifically, decision variables are used as structural variables, and constraint-based scheduling strategies are used as scheduling variables. For the constructed microgrid cluster multi-objective optimization model, an improved multi-objective dragonfly algorithm is adopted. By using structural variables and scheduling variables, the multi-objective optimization model of the microgrid cluster is jointly optimized by simulating the behavior rules of the dragonfly swarm.

[0057] To overcome the problems of premature convergence and local optima that traditional Dragonfly Algorithm (DA) is prone to in solving complex high-dimensional nonlinear optimization problems, an improved version of the Dragonfly Algorithm is adopted using chaotic mapping and nonlinear inertial weighting strategies. This enhances the global search capability and convergence accuracy of the optimization algorithm and balances various objectives more efficiently.

[0058] An improved multi-objective dragonfly algorithm is used to efficiently solve the microgrid group collaborative optimization problem. Chaotic mapping is used to initialize the dragonfly population to enhance population diversity and avoid the algorithm getting trapped in local optima too early. Nonlinear inertial weights are used to dynamically adjust the update step size of individual dragonflies, enabling the algorithm to quickly explore the solution space in the early stage and gradually converge to the optimal solution in the later stage.

[0059] Specifically, the dragonfly's position update process is as follows: ; ; ; in: and They represent dragonflies. Current position and next position; and They represent dragonflies. Current step size and next step size; They represent dragonflies. Functions for separation behavior, alignment behavior, aggregation behavior, foraging behavior, and enemy avoidance behavior. These represent the weight coefficients of the functions for separation behavior, alignment behavior, clustering behavior, foraging behavior, and enemy avoidance behavior, respectively. Indicates inertia weight; and These represent the upper and lower limits of the inertia weight, respectively; Represents a constant; This indicates the maximum number of iterations.

[0060] During the dragonfly position update process, based on the heterogeneity of variables, the control of structural variables and scheduling variables is handled in a domain-specific manner. Specifically, separation and aggregation behavior functions control structural variables, i.e., the connection state between power grids, while alignment, foraging, and enemy avoidance behavior functions control and guide scheduling variables. This achieves coordinated decision-making and unified updates of structure and operation, thereby significantly improving the coordinated scheduling effect of microgrid groups under complex constraints. This is a variable-behavior integrated design mechanism, realizing a structural upgrade of the dragonfly algorithm from "unified behavior to adaptive mapping behavior".

[0061] Specifically, structural variables (such as connection state) and scheduling variables (such as power P and energy storage SOC) are packaged into a single "individual code" and used as the optimization object of the Dragonfly algorithm. However, during the optimization process, not all variables are treated equally. Instead, a "domain-specific optimization" strategy is adopted: for structural variables, only behaviors such as "separation" and "aggregation" are used for adjustment; for scheduling variables, "foraging", "avoiding enemies", and "alignment" are used for optimization. Finally, these two parts are merged to form a new overall solution, which is then fed into a multi-objective function (economy, curtailment rate, carbon emissions, etc.) to calculate the score.

[0062] The dynamic swarming mechanism is no longer just a modeling input, but participates in the optimization process as an "adjustable decision variable"; and through the above-mentioned structure-behavior coordination mechanism, it guides and evolves, solving the problem that the optimization of "swarm structure-running parameters" in traditional scheduling algorithms is difficult to model and converge in a unified manner.

[0063] Specifically, based on the improved multi-objective dragonfly algorithm, the collaborative optimization of swarm structure and scheduling scheme and the search for the optimal solution set are achieved. The main steps are as follows: (1) Collect microgrid group operation data, including distributed power sources, loads, energy storage data, and electricity price information; (2) Set optimization objectives and decision variables. Each objective is combined by weighting to form the overall objective, including minimizing operating costs, carbon emission costs and new energy curtailment costs. At the same time, network structure is optimized through decision variables. (3) Set the basic parameters of the multi-objective dragonfly algorithm, including the number of dragonflies and the upper and lower limits of the inertial weight, and initialize the dragonfly population through chaotic mapping; (4) Simulate the behavior rules of dragonflies, perform behaviors such as separation, alignment, gathering, foraging and avoiding enemies, classify and use the behavior functions according to variable types, calculate the corresponding movement step for each variable, and then splice them to form the total individual change, uniformly update the position of the entire dragonfly individual, and calculate the fitness of the dragonfly individual to ensure the balance of each objective in multi-objective optimization. (5) Update the nonlinear inertia weight and the weight coefficients of the five behaviors to realize the position update of the dragonfly; (6) Update the dragonfly's position through multiple iterations, and determine whether the optimal solution has been reached by combining the maximum number of iterations and the convergence criterion; (7) After multiple iterations, the solution with the highest fitness is selected as the optimal solution, thus obtaining the solution of the objective function that minimizes the total cost, thereby realizing the microgrid group scheduling optimization.

[0064] Based on the differences in the types of optimization variables, this invention constructs a behavior-variable mapping mechanism, which allocates the typical behavior functions in the multi-objective dragonfly algorithm according to the characteristics of the variables, and guides the structural variables and scheduling variables to be updated in different search subspaces respectively, thereby enhancing the pertinence of behavior control and the stability of optimization convergence, and realizing the unified evolution of individual dragonflies by combining step sizes.

[0065] Step four: Output Pareto optimized scheduling solution sets covering different operating preferences to obtain a multi-energy flow coordinated operation scheme for low-carbon economy of microgrid groups, so as to maximize the consumption of renewable energy while reducing operating costs.

[0066] The improved multi-objective dragonfly algorithm optimizes the scheduling and operation of microgrid groups. Individual dragonflies can adjust their flight strategies in a timely manner according to the changes in the fitness function in a dynamic environment, thereby improving the global optimization capability and outputting a Pareto scheduling solution set covering different operating preferences. Guided by minimizing the total cost objective function, the optimal microgrid group scheduling strategy is obtained.

[0067] Example 2 This invention also proposes a microgrid group operation optimization system based on the multi-objective dragonfly algorithm, comprising: The microgrid cluster system modeling module constructs a system model of a microgrid cluster, which includes multiple interconnected microgrids. The connection status of the multiple microgrids is used as a decision variable to characterize the interconnection mode between the microgrids. The multi-objective optimization modeling module establishes a multi-objective optimization model that comprehensively considers minimizing operating costs, carbon emissions, and renewable energy curtailment losses, and designs operating constraints including power balance, energy storage limitations, and power exchange capacity. The structure-scheduling joint optimization module combines decision variables and constraints, and adopts an improved multi-objective dragonfly algorithm to design a structure-scheduling joint optimization strategy, thereby achieving coordinated optimization of group structure and scheduling strategy and search for the optimal solution set. The optimized solution set output module outputs Pareto optimized scheduling solutions covering different operating preferences, resulting in a multi-energy flow coordinated operation scheme for low-carbon economy of microgrid groups, which maximizes the consumption of renewable energy while reducing operating costs.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A microgrid group operation optimization method based on multi-objective dragonfly algorithm, characterized in that, include: Step 1: Construct a system model of a microgrid cluster, which includes multiple microgrids; The connection status of the multiple microgrids is used as a decision variable to represent the interconnection mode between the microgrids; a dynamic grouping mechanism is introduced to adjust the connection relationship of the multiple microgrids according to actual needs. Step 2: Establish a multi-objective optimization model that comprehensively considers minimizing operating costs, carbon emissions, and renewable energy curtailment losses, and design operating constraints including power balance, energy storage limitations, and power exchange capacity. Step 3: Combining the decision variables and constraints, design a structure-scheduling joint optimization strategy to collaboratively optimize the microgrid group structure and scheduling strategy; Step four: Based on the optimization results, output the Pareto optimized scheduling solution set covering different operating preferences to obtain the multi-energy flow coordinated operation scheme of the microgrid group for low-carbon economy.

2. The microgrid group operation optimization method based on multi-objective dragonfly algorithm according to claim 1, characterized in that: In step one, binary variables are used. To indicate whether microgrids are connected, i.e. ;when Represents the microgrid at time t and microgrids There are connections between them; Represents the microgrid at time t and microgrids No connection.

3. The microgrid group operation optimization method based on multi-objective dragonfly algorithm according to claim 1, characterized in that: In step one, the dynamic grouping mechanism allows the microgrid to adaptively adjust its networking mode within the scheduling cycle or according to changes in system load; The dynamic grouping mechanism is specifically expressed as follows: under the premise of satisfying the transmission capacity of the connection lines, the microgrid... and microgrids The connection status adjustment process is as follows: ； ； ， ； in, microgrid and microgrids The complementary needs between them serve as the basis for determining the connection; Indicates the connection hysteresis compensation threshold; Represents a binary variable, indicating whether microgrids are connected; This represents a binary variable adjusted through a dynamic grouping mechanism. This indicates the line loss caused by switching power.

4. The microgrid group operation optimization method based on multi-objective dragonfly algorithm according to claim 2, characterized in that: In step one, when the microgrid and microgrids Interconnection, under the constraint of meeting the transmission capacity of the connection lines, microgrids and The calculation of power exchange between them is expressed by the following equation: ； in: microgrid and The actual amount of power exchanged between them; Represents a binary variable, indicating whether microgrids are connected; and They represent microgrids and The output power; and They represent microgrids and The electrical load; Indicates the maximum power transmission capacity of the connecting line; This indicates the line loss caused by switching power.

5. The microgrid group operation optimization method based on multi-objective dragonfly algorithm according to claim 1, characterized in that: In step three, the structure-scheduling joint optimization strategy adopts an improved multi-objective dragonfly algorithm, which uses structural variables and scheduling variables to jointly optimize the multi-objective optimization model of the microgrid cluster by simulating the behavior rules of the dragonfly swarm. The decision variables are used as structural variables, and the constraint-based scheduling strategy is used as the scheduling variable.

6. The microgrid group operation optimization method based on multi-objective dragonfly algorithm according to claim 5, characterized in that: In step three, the improved multi-objective dragonfly algorithm is improved by using chaotic mapping and nonlinear inertial weighting strategies. The dragonfly population is initialized using chaotic mapping, and the update step size of individual dragonflies is adjusted and the position of the dragonflies is updated using nonlinear inertial weighting.

7. The microgrid group operation optimization method based on multi-objective dragonfly algorithm according to claim 5, characterized in that: In step three, the updating of the dragonfly's position is related to the dragonfly's separation behavior, alignment behavior, gathering behavior, foraging behavior, and enemy avoidance behavior. Based on the heterogeneity of structural variables and scheduling variables, the control of structural variables and scheduling variables is performed in a domain-specific manner.

8. The microgrid group operation optimization method based on multi-objective dragonfly algorithm according to claim 7, characterized in that: In step three, the structural variables and scheduling variables are packaged into a single individual code, which is then used as the object of optimization by the Dragonfly algorithm. In the optimization process, domain-specific processing refers to adjusting structural variables using separation and aggregation behaviors. For runtime variables, optimization is achieved using alignment behavior, foraging behavior, and enemy avoidance behavior; The results of the domain optimization are merged to form the overall individual changes, and the position of the entire dragonfly is updated uniformly.

9. A microgrid cluster operation optimization system based on a multi-objective dragonfly algorithm using the method described in any one of claims 1-8, comprising a microgrid cluster system modeling module, a multi-objective optimization modeling module, a structure-scheduling joint optimization module, and an optimization solution set output module, characterized in that: The microgrid cluster system modeling module constructs a system model of a microgrid cluster, which includes multiple microgrids; the connection status of the multiple microgrids is used as a decision variable to characterize the interconnection mode between the microgrids. The multi-objective optimization modeling module establishes a multi-objective optimization model that comprehensively considers minimizing operating costs, carbon emissions, and renewable energy curtailment losses, and designs operating constraints including power balance, energy storage limitations, and power exchange capacity. The structure-scheduling joint optimization module combines decision variables and constraints, and adopts an improved multi-objective dragonfly algorithm to design a structure-scheduling joint optimization strategy, thereby achieving coordinated optimization of group structure and scheduling strategy and search for the optimal solution set. The optimized solution set output module outputs Pareto optimized scheduling solutions covering different operating preferences, resulting in a multi-energy flow coordinated operation scheme for low-carbon economy of microgrid groups.

10. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-8.

Images