Micro-grid energy management method based on hierarchical genetic algorithm

Abstract

The invention provides a micro-grid energy management method based on a hierarchical genetic algorithm. The micro-grid energy management method comprises the steps that whether a gas turbine of a nearby micro-grid is activated or an energy storage battery of the micro-grid is utilized is determined according to the power supply condition; whether unused micro-grid resources exist or not is checked, and gas turbines of other micro-grids continue to be used; if the micro-grid still has insufficient power supply after using the gas turbine, entering a standby process, and using energy storage batteries of other micro-grids; energy storage batteries of other micro-grids are continuously used according to requirements until all power requirements are met; repeatedly executing each micro-grid until the power supply and demand balance of all micro-grids are processed; and updating the current time to the next measurement moment, and circularly executing in a 24-hour operation period. According to the method, two layers of optimization models are established, the upper layer aims at reducing power loss, improving voltage distribution and smoothing power fluctuation of power distribution network operation, and the lower layer aims at reducing MG operation cost and pollution emission.

Description

Technical Field

[0001] The present invention relates to the technical field of power system management, and in particular to a microgrid energy management method based on a hierarchical genetic algorithm. Background Art

[0002] In recent years, with the growth of energy demand and the urgent need for sustainable energy supply, the integration of distributed energy resources (DERs) has gradually become an effective way to reconstruct the existing distribution network and improve the reliability of energy supply. Microgrids (MGs), as an advanced technology integrating distributed generators, loads and energy storage systems, have shown their potential advantages in many aspects, including but not limited to the improvement of system operation efficiency, reliability, economic benefits and environmental friendliness.

[0003] Despite this, the integration and operation of microgrids still face many challenges. For example, microgrid management in existing technologies usually fails to fully consider the overall optimization of the system, resulting in low resource utilization, insufficient cost savings potential, and insufficient environmental impact mitigation. In addition, although existing solutions have made certain contributions to improving economic benefits in theory, how to comprehensively consider and optimize the energy management of microgrids in actual applications to maximize cost-effectiveness, especially in complex distributed systems with multiple microgrids interconnected, is still an unresolved problem.

[0004] To this end, the present invention provides a clustered microgrid energy management method based on a hierarchical genetic algorithm for microgrids, which can not only take into account the efficiency and economy of a single microgrid, but also achieve optimization at a broader system level to meet the growing requirements of modern power distribution systems for flexibility and reliability. Summary of the invention

[0005] In order to overcome the shortcomings of the existing technology, the present invention proposes a microgrid energy management method based on a hierarchical genetic algorithm to optimize the operation of the microgrid and the distribution network, achieve the dual goals of economic benefits and environmental protection, and enhance the stability and responsiveness of the power system. It not only improves energy efficiency, but also promotes a wider range of sustainable development and the application of smart grid technology.

[0006] To achieve the above object, the present invention provides a microgrid energy management method based on a hierarchical genetic algorithm, comprising:

[0007] Step S1: At a set time point, the power supply of the i-th microgrid node is evaluated, and according to the power supply situation, it is decided whether to activate the gas turbine of the nearby microgrid or use the energy storage battery of the microgrid, and the power generation or charging and discharging amount is calculated using a dual-objective genetic algorithm;

[0008] Step S2: Check whether there are any unused microgrid resources, and re-evaluate the power status of the i-th microgrid. If necessary, continue to use the gas turbines of other microgrids until the power demand is met or all resources are used up;

[0009] Step S3: If the i-th microgrid still has insufficient power supply after using the gas turbine, it enters the backup process, uses the energy storage batteries of other microgrids, and uses the dual-objective genetic algorithm to recalculate the required power generation;

[0010] Step S4: After activating the energy storage battery, re-evaluate the power supply status and continue to use the energy storage batteries of other microgrids as needed until all power needs are met;

[0011] Step S5: Repeat steps S1 to S4 for each microgrid until the power supply and demand balance of all microgrids is processed;

[0012] Step S6: Update the current time to the next measurement time, and cyclically execute steps S1 to S5 within a 24-hour operation cycle to maintain continuous power supply and management.

[0013] Further, step S1 includes:

[0014] Step S11: at a set time point, monitoring the power supply status of the node where the i-th microgrid is located;

[0015] Step S12: judging according to the power supply situation, if the power supply is insufficient, executing step S13, if the power supply is sufficient, executing step S14;

[0016] Step S13: activating gas turbines of adjacent microgrids in order of priority to support power demand;

[0017] Step S14: activating the energy storage battery of the microgrid to maintain load balance;

[0018] Step S15: Calculate the required power generation or charge and discharge amount using the improved dual-objective genetic algorithm.

[0019] Further, step S2 includes:

[0020] Step S21: Check whether there are any unused microgrid resources;

[0021] Step S22: re-evaluate the power status of the i-th microgrid;

[0022] Step S23: If the i-th microgrid still has insufficient power supply, continue to activate gas turbines of more microgrids;

[0023] Step S24: Check whether the responsiveness reserves of all microgrids have been fully called. If so, end this step; if not, return to step S21.

[0024] Further, step S3 includes:

[0025] Step S31: confirm whether the i-th microgrid still has insufficient power supply after using the gas turbine;

[0026] Step S32: If the power supply is insufficient, activate the energy storage batteries in other microgrids to provide support;

[0027] Step S33: Use the improved dual-objective genetic algorithm to recalculate the required power generation to ensure power supply.

[0028] Further, step S4 includes:

[0029] Step S41: after activating the energy storage battery, re-evaluate the power supply status of the i-th microgrid;

[0030] Step S42: If the power is still insufficient, continue to activate the energy storage battery of the next microgrid;

[0031] Step S43: Repeat the evaluation and activation process until the power demand is met.

[0032] Further, step S5 includes:

[0033] Step S51: Select the next microgrid for evaluation;

[0034] Step S52: Repeat steps S1 to S4 for the selected microgrid;

[0035] Step S53: Check whether all microgrids have been evaluated and processed. If not, return to step S51; if completed, end this step.

[0036] Further, step S6 includes:

[0037] Step S61: At the end of the cycle, update the time to the next monitoring moment;

[0038] Step S62: repeating steps S1 to S5 until the entire 24-hour operation cycle is covered;

[0039] Step S63: Monitor and evaluate the performance and power supply status of the entire system and make necessary adjustments.

[0040] Further, in step S13, the gas turbines of other microgrids are activated in sequence, wherein the sequence is determined by the priority. The microgrids with remaining gas turbine reserves are sorted according to the size of the remaining reserves, and the larger the size, the higher the priority. The remaining reserves are calculated as follows:

[0041] N i =P i -P i load (1)

[0042] Among them, N i is the remaining reserve of the i-th microgrid, P i represents the total power generation of the i-th microgrid, P i load Represents the load of the node where the i-th microgrid is located.

[0043] Introduce binary variable array X f,t =[x h,t ...x H,t ],Y f,t =[y h,t ...y H,t ] is used to describe the priority of the microgrid. Each microgrid has its own binary variable as a trigger. When the trigger condition is met, the trigger is set to 1, indicating that the corresponding microgrid reserve is executed. The trigger conditions are: 1) The previous trigger is 1; if the current trigger has the highest priority, this condition can be ignored; 2) After the previous power supply microgrid is executed, the disadvantage microgrid still needs power support. The mathematical expression is as follows:

[0044]

[0045]

[0046] or

[0047]

[0048] Where: f∈F is the set of all microgrids; h∈H f A collection of microgrids that can provide power assistance, and Microgrid MG h The remaining dispatchable battery storage (SB) and gas turbine (MT) capacity, X f,t and Y f,t Yes H f ×1 matrix. h,t and h,t It controls MG h The trigger for gas turbine startup, α h,t , β h,tThey are The maximum allowable remaining dispatchable capacity, α is the limiting parameter, For MG h The gas turbine power generation capacity, For MG h The transition charge state of SB.

[0049] Furthermore, the dual-objective genetic algorithm in step S15 comprises the following steps:

[0050] 1) Initialize parameters, set chromosomes, population size, and the number of iterations to 1, and read data;

[0051] 2) Randomly create individuals and check whether they meet the feasibility requirements. Keep the individuals that meet the feasibility requirements until the number reaches the set population size.

[0052] 3) Calculate the fitness value according to the upper layer objective function;

[0053] 4) Select individuals with higher fitness in the current population;

[0054] 5) Cross the genes in the individuals, check the feasibility of the generated individuals, eliminate the individuals that lead to infeasible solutions, and continue crossover and mutation until the next generation of population is formed;

[0055] 6) If the convergence condition is met, the optimal solution is returned and the algorithm ends; otherwise, go to 3);

[0056] 7) The energy management system executes energy management and scheduling strategies according to the optimal solution.

[0057] Further, in 2), the step of checking whether the individual satisfies feasibility is as follows: each exchange variable of the upper individual Substitute into the lower layer The improved hierarchical genetic algorithm (HGA) is used to calculate the lower model to check whether the lower optimization problem has a solution. If so, the feasibility is met, otherwise it does not meet the feasibility.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. The present invention provides a microgrid energy management method based on a hierarchical genetic algorithm. By minimizing power fluctuations and power losses, this method can smooth the power fluctuations in the distribution network, thereby improving the overall efficiency and reliability of the power system. At the same time, improving voltage distribution helps maintain the network voltage within a stable range, avoiding equipment damage and operation risks caused by excessively high or low voltage.

[0060] 2. The present invention provides a microgrid energy management method based on a hierarchical genetic algorithm, the lower level goal of which focuses on reducing the operating costs and pollution emissions of the microgrid (MG). This is achieved by optimizing the use of distributed energy and energy storage facilities, and rationally dispatching controllable energy, which helps to reduce the economic burden and environmental pollution caused by the use of conventional energy.

[0061] 3. The present invention provides a microgrid energy management method based on a hierarchical genetic algorithm. The concepts of response reserve and remaining dispatchable capacity introduced in the model make the system more flexible and responsive in the face of demand fluctuations and emergencies. This design can quickly adjust power generation and load to ensure grid stability while reducing possible system pressure.

[0062] 4. The present invention provides a microgrid energy management method based on hierarchical genetic algorithm, which uses interactive game matrix (IGM) and game theory principles to mathematically model the operations between microgrid and distribution network, and promotes the cooperation and coordination of all parties in resource and operation management. This method not only optimizes the interests of each party, but also ensures the optimal operation of the overall system.

[0063] 5. The present invention provides a microgrid energy management method based on a hierarchical genetic algorithm. The improved hierarchical genetic algorithm provides an effective tool to deal with such multi-level, multi-objective optimization problems. The algorithm can effectively find the optimal solution that meets the upper and lower level objectives, while dealing with complex constraints and dynamic changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0065] Figure 1 It is a schematic diagram of the steps of the present invention

[0066] Figure 2 This is a schematic diagram of the distribution network management system. The distribution network system consists of nodes, microgrids, transformers, transmission lines and energy management systems (EMS). MG1, MG2, and MG3 represent different microgrids, each of which consists of power generation equipment and energy storage equipment; the feeder is represented by a solid line, which indicates the line from the root node to the load node; the connecting branch line is represented by a dotted line, which indicates the line from the microgrid to the load node; the EMS can monitor and manage each node to achieve a balance between power supply and demand and optimize operating costs;

[0067] Figure 3 Improved hierarchical genetic algorithm flow chart;

[0068] Figure 4 is the load curve of the distribution network. DETAILED DESCRIPTION

[0069] The technical solution of the present invention will be more clearly and completely explained below through description of preferred embodiments of the present invention in combination with the accompanying drawings.

[0070] Terminology explanation:

[0071] 1. Microgrid (MG): A small-scale grid system, including power generation, distribution, storage and consumption facilities, which can operate independently of the traditional grid or be connected to it. It usually includes renewable energy sources such as solar and wind power.

[0072] 2. Bi-objective Genetic Algorithm: A search algorithm used to optimize two objectives at the same time, often used to solve multi-objective optimization problems. It simulates the process of natural selection and genetics to generate optimal or near-optimal solutions.

[0073] 3. Responsive Reserve: refers to backup power resources that can be quickly called upon to maintain grid stability when demand increases sharply or when the main power source fails.

[0074] 4. Energy Storage: Technology or facilities used to store energy for later use, such as battery energy storage systems, which can release electricity during peak demand to balance supply and demand.

[0075] 5. Controllable Sources: refers to energy facilities that can adjust their output according to grid demand, such as gas generators, as opposed to unpredictable renewable energy sources (such as wind and solar power).

[0076] 6. Interactive Game Matrix (IGM): A mathematical tool used to describe and analyze the strategic choices of multiple participants and their interactions. It is often used in economic and social science research, as well as in game theory analysis of complex systems.

[0077] 7. Game Theory: A branch of economics that studies the interaction strategies between decision makers (players) with conflicting and cooperative motivations.

[0078] 8. Hierarchical Genetic Algorithm: A variant of the genetic algorithm that improves the algorithm's search capability by setting up multi-level genetic operations. It is particularly suitable for solving optimization problems with multi-level decision-making structures.

[0079] 9. Voltage Deviation: Describes the difference between the voltage at a point in the power grid and its nominal (or reference) value. Voltage stability is an important indicator of the health of the power system.

[0080] 10. Power Loss: Energy loss caused by factors such as equipment resistance during power transmission and distribution. Reducing power loss is a key goal to improve grid efficiency.

[0081] 11. Operational Cost: All costs involved in maintaining the operation of the power system, including fuel, maintenance, equipment depreciation, etc.

[0082] 12. Emissions: refers to pollutants produced during the power generation process, such as carbon dioxide, nitrogen oxides, etc., which usually come from power plants burning fossil fuels.

[0083] like Figure 1 As shown, the present invention is specifically:

[0084] Step S1: Evaluate the power supply of the microgrid node and decide whether to activate the gas turbine of the nearby microgrid or use the energy storage battery of the microgrid according to the power supply situation, and use the dual-objective genetic algorithm to calculate the power generation or charging and discharging amount;

[0085] Step S2: Check whether there are any unused microgrid resources, re-evaluate the power status of the microgrid, and continue to use the gas turbines of other microgrids until the power demand is met or all resources are used up;

[0086] Step S3: If the microgrid still has insufficient power supply after using the gas turbine, it enters the backup process, uses the energy storage batteries of other microgrids, and uses the dual-objective genetic algorithm to recalculate the required power generation;

[0087] Step S4: After activating the energy storage battery, re-evaluate the power supply status and continue to use the energy storage batteries of other microgrids as needed until all power needs are met;

[0088] Step S5: Repeat steps S1 to S4 for each microgrid until the power supply and demand balance of all microgrids is processed;

[0089] Step S6: Update the current time to the next measurement time, and cyclically execute steps S1 to S5 within a 24-hour operation cycle to maintain continuous power supply and management.

[0090] As a specific embodiment, at time t, the power supply status of the node where the i-th microgrid is located is evaluated. If there is insufficient power supply, other microgrids are called to provide power support, the gas turbines in each microgrid are activated in sequence, and the power generation of each microgrid is calculated using an improved dual-objective genetic algorithm (considering power quality and cost optimization); if the power supply is sufficient, the energy storage battery is activated to maintain the load balance of the system, and the charge and discharge amount is calculated using the same algorithm. Then go to step S2.

[0091] Among them, the gas turbines of other microgrids are activated in order, and the order is determined by the priority. The microgrids with remaining gas turbine reserves are sorted according to the size of the remaining reserves. The larger the priority, the higher the priority. The remaining reserves are calculated as follows:

[0092] N i =P i -P i load (1)

[0093] Among them, N i is the remaining reserve of the i-th microgrid, P i represents the total power generation of the i-th microgrid, P i load Represents the load of the node where the i-th microgrid is located.

[0094] Introduce binary variable array X f,t =[x h,t ...x H,t ],Y f,t =[y h,t ...y H,t ] is used to describe the priority of the microgrid. Each microgrid has its own binary variable as a trigger. When the trigger condition is met, the trigger is set to 1, indicating that the corresponding microgrid reserve is executed. The trigger conditions are: 1) The previous trigger is 1; if the current trigger has the highest priority, this condition can be ignored; 2) After the previous power supply microgrid is executed, the disadvantage microgrid still needs power support. The mathematical expression is as follows:

[0095]

[0096]

[0097] or

[0098]

[0099] Where: f∈F is the set of all microgrids; h∈H f A collection of microgrids that can provide power assistance, and Microgrid MG h The remaining dispatchable battery storage (SB) and gas turbine (MT) capacity, X f,t and Y f,t Yes H f ×1 matrix. h,t and h,t It controls MG h The trigger for gas turbine startup, α h,t , β h,t They are The maximum allowable remaining dispatchable capacity, α is the limiting parameter, For MG h The gas turbine power generation capacity, For MG h The transition charge state of SB.

[0100] Bi-objective genetic algorithm, such as Figure 3 As shown, the steps are:

[0101] 1) Initialize parameters, set chromosomes, population size, and set the number of iterations to 1, read data, including node voltage, node power, node load, node load, and predicted values ​​of wind power generation and photovoltaic power generation in the microgrid;

[0102] 2) Randomly create individuals and check whether the individuals meet the feasibility requirements. Keep the individuals that meet the feasibility requirements until the number reaches the set population size. The checking steps are: Substitute into the lower layer The improved hierarchical genetic algorithm (HGA) is used to calculate the lower model to check whether the lower optimization problem has a solution. If so, the feasibility is met, otherwise it does not meet the feasibility.

[0103] The improved hierarchical genetic algorithm calculates the lower model, which is characterized in that the mathematical expression of the lower model is:

[0104] minf down =f1+f2 (10)

[0105]

[0106] st

[0107]

[0108] in:

[0109]

[0110] Where: are fuel and operating costs respectively; is the variable in the interactive game matrix; C f,t for Cost coefficient of For microgrid MG f Power exchange with the distribution network, q is the electricity price; They represent the pollutant emissions and the corresponding costs respectively; Representing Microgrid MG f The power generation of distributed energy in , s=1,2,3,4∈S f Represents MG f In the middle, there are gas turbines, energy storage batteries, photovoltaics, and wind energy; F is the collection of microgrids; is the excess charge state of SB, E f,t is the SOC of SB; R up ,R down is the ramp constraint of the gas turbine; It is the extra power that the microgrid can provide in addition to meeting the normal load demand. f1 represents the total operating cost of all microgrids, and f2 represents the pollution emission penalty;

[0111] In the improved genetic algorithm, the genes in the chromosome are divided into four layers, namely: high-level control genes a f,t ,b f,t , controlling gene x h,t ,y h,t , parameter gene Common genes Parameter genes are genes affected by control genes, and common genes are genes not affected by control genes. f,t ,b f,t Used to control the way the microgrid provides assistance, i.e. using gas turbines or energy storage batteries, x h,t ,y h,t Used to control which microgrid provides assistance; the encoding format of the chromosome is During chromosome initialization, a f,t ,b f,t ,x h,t ,y h,t Determined by the following formula, other genes are randomly generated.

[0112]

[0113] The upper objective function has the following mathematical model:

[0114] f up=μ1F1+μ2F2+μ3F3 (21)

[0115]

[0116] st

[0117]

[0118] Where: V i,t is the voltage of node i; is the active transmission power between the node and the microgrid, is the reactive transmission power between the node and the microgrid; η is the degree of power exchange; P t LS Power loss in the distribution network; G i,j , B i,j ,θ i,j,t Represent the conductance, susceptance and phase difference in lines i and j respectively; R t It is the original responsiveness reserve of the distribution network; The original responsiveness reserve for all microgrids; P gmax ,Q gmax is the maximum output of the gas turbine and the reactive power compensator. F1 represents the voltage deviation of all nodes, F2 represents the fluctuation of power transmission between the distribution network and the microgrid, and F3 represents the total power loss of the distribution network.

[0119] 3) Calculate the fitness value according to the upper layer objective function;

[0120] 4) Select individuals with higher fitness in the current population;

[0121] 5) Cross the genes in the individuals, check the feasibility of the generated individuals, eliminate the individuals that lead to infeasible solutions, and continue crossover and mutation until the next generation of population is formed;

[0122] 6) If the convergence condition is met, the optimal solution is returned and the algorithm ends; otherwise, go to step 3);

[0123] 7) The energy management system executes energy management and scheduling strategies according to the optimal solution.

[0124] Evaluate whether there are still unparticipated microgrids and check whether the i-th microgrid is still under-powered. If so, continue to activate the gas turbines of the remaining microgrids until the power demand is met or the responsiveness reserves of all microgrids are called. Then go to step S3.

[0125] If the i-th microgrid still has insufficient power supply after the gas turbine is called, the backup process is started: the energy storage batteries in other microgrids are activated, and the power generation is recalculated using the improved dual-objective genetic algorithm. If the power supply is sufficient, the energy storage battery activation is skipped and step S5 is directly entered.

[0126] After activating the energy storage battery, re-evaluate the power supply status of the i-th microgrid. If it is still insufficient, continue to activate the energy storage battery of the next microgrid until all the energy storage batteries of the microgrid are used or the power demand is met. After completion, go to step S5.

[0127] The next microgrid is subjected to the operations of S1 to S4 until all microgrids are checked. After the evaluation and scheduling of all microgrids are completed, the process proceeds to step S6.

[0128] Update time t to the next moment, and repeat steps S1 to S5 for all microgrids until a complete operation cycle (24 hours) is completed.

[0129] As a specific implementation case, Figure 2 As shown, this example uses a modified IEEE14 bus distribution system with three microgrids. For this test system, the voltage reference is 10.5kV, the total active load is 2870kW, and the total reactive load is 775kVAR. 230kW generators are connected at nodes 6, 11, and 12. -100~300kVAR static var compensation (SVC) is connected at nodes 7 and 13. It is set to 8kW. Table 1 summarizes the parameters of the microgrid. q is set to 4.75￥ / kWh. Figure 4 is the load of the entire distribution network.

[0130] Table 1 Parameters of microgrid

[0131]

[0132] 1. Load 24-hour forecast data of node load, node power, voltage, photovoltaic and wind power generation;

[0133] 2. When t = 1, determine whether the microgrid is underpowered. If so, a f,t =1, activate the generator to supply power, according to the control variable X f,t =[x h,t ...x H,t ] to decide which micro-grid to provide power, and the power provided is determined according to the dual-objective genetic algorithm. Until there is no power shortage or all microgrids have been supplied, if the power supply is sufficient, then a f,t =0;

[0134] 3. Determine whether the microgrid is still underpowered. If so, a f,t =0,b f,t =1, activate the energy storage battery to supply power, according to the control variable Y f,t =[y h,t ...y H,t] to decide which micro-grid to choose for power supply, and the power provided is determined according to the dual-objective genetic algorithm. Until there is no power shortage or all microgrids have been fully supplied;

[0135] 4. Repeat steps 1 to 3 for the next microgrid until all microgrids are operated.

[0136] 5. After each traversal of the microgrid, t+1, repeat 1 to 4 until t=24.

[0137] Figure 4 The load curve of the distribution network. Through the analysis of the curve, we can see that the load curve shows the effect of peak shaving and valley filling after energy management, that is, during the peak load period, such as 10 am to 8 pm, the load is effectively reduced; while during the valley period, such as 1 am to 5 am at night, the load is increased. This change shows that the patented method effectively balances the load of the distribution network through the energy management of the microgrid, enhances the stability of the power grid, and may reduce the overall cost of electricity.

[0138] The above specific implementations are only descriptions of the preferred implementations of the present invention, and do not limit the protection scope of the present invention. Without departing from the design concept and spirit of the present invention, various modifications, substitutions and improvements made by ordinary technicians in this field to the technical solution of the present invention based on the text description and drawings provided by the present invention should all fall within the protection scope of the present invention. The protection scope of the present invention is determined by the claims.

Claims

1. A microgrid energy management method based on hierarchical genetic algorithm, characterized in that: include: Step S1: Evaluate the power supply of the microgrid node and decide whether to activate the gas turbine of the nearby microgrid or use the energy storage battery of the microgrid according to the power supply situation, and use the dual-objective genetic algorithm to calculate the power generation or charging and discharging amount; Step S2: Check whether there are any unused microgrid resources, re-evaluate the power status of the microgrid, and continue to use the gas turbines of other microgrids until the power demand is met or all resources are used up; Step S3: If the microgrid still has insufficient power supply after using the gas turbine, it enters the backup process, uses the energy storage batteries of other microgrids, and uses the dual-objective genetic algorithm to recalculate the required power generation; Step S4: After activating the energy storage battery, re-evaluate the power supply status and continue to use the energy storage batteries of other microgrids as needed until all power needs are met; Step S5: Repeat steps S1 to S4 for each microgrid until the power supply and demand balance of all microgrids is processed; Step S6: Update the current time to the next measurement time, and cyclically execute steps S1 to S5 within a 24-hour operation cycle to maintain continuous power supply and management.

2. The microgrid energy management method based on hierarchical genetic algorithm according to claim 1 is characterized in that: Step S1 includes: Step S11: at a set time point, monitoring the power supply status of the node where the i-th microgrid is located; Step S12: judging according to the power supply situation, if the power supply is insufficient, executing step S13, if the power supply is sufficient, executing step S14; Step S13: activating gas turbines of adjacent microgrids in order of priority to support power demand; Step S14: activating the energy storage battery of the microgrid to maintain load balance; Step S15: Calculate the required power generation or charge and discharge amount using the improved dual-objective genetic algorithm.

3. The microgrid energy management method based on hierarchical genetic algorithm according to claim 1 is characterized in that: Step S2 includes: Step S21: Check whether there are any unused microgrid resources; Step S22: re-evaluate the power status of the microgrid; Step S23: If the microgrid still has insufficient power supply, continue to activate more gas turbines in the microgrid; Step S24: Check whether the responsiveness reserves of all microgrids have been fully called. If so, end this step; if not, return to step S21.

4. The microgrid energy management method based on hierarchical genetic algorithm according to claim 1 is characterized in that: Step S3 includes: Step S31: confirm whether the microgrid still has insufficient power supply after using the gas turbine; Step S32: If the power supply is insufficient, activate the energy storage batteries in other microgrids to provide support; Step S33: Use the improved dual-objective genetic algorithm to recalculate the required power generation to ensure power supply.

5. The microgrid energy management method based on hierarchical genetic algorithm according to claim 1 is characterized in that: Step S4 includes: Step S41: after activating the energy storage battery, re-evaluate the power supply status of the microgrid; Step S42: If the power is still insufficient, continue to activate the energy storage battery of the next microgrid; Step S43: Repeat the evaluation and activation process until the power demand is met.

6. A microgrid energy management method based on hierarchical genetic algorithm according to claim 1, characterized in that: Step S5 includes: Step S51: Select the next microgrid for evaluation; Step S52: Repeat steps S1 to S4 for the selected microgrid; Step S53: Check whether all microgrids have been evaluated and processed. If not, return to step S51; if completed, end this step.

7. A microgrid energy management method based on hierarchical genetic algorithm according to claim 1, characterized in that: Step S6 includes: Step S61: at the end of the cycle, update the time to the next monitoring moment; Step S62: repeating steps S1 to S5 until the entire 24-hour operation cycle is covered; Step S63: Monitor and evaluate the performance and power supply status of the entire system and make necessary adjustments.

8. The microgrid energy management method based on hierarchical genetic algorithm according to claim 1 is characterized in that: In step S13, the gas turbines of other microgrids are activated in sequence, wherein the sequence is determined by the priority. The microgrids with remaining gas turbine reserves are sorted according to the size of the remaining reserves. The larger the size, the higher the priority. The remaining reserves are calculated as follows: N i =P i -P i load (1) Among them, N i is the remaining reserve of the i-th microgrid, P i represents the total power generation of the i-th microgrid, P i load represents the load of the node where the i-th microgrid is located; Introduce binary variable array X f,t =[x h,t ...x H,t ],Y f,t =[y h,t ...y H,t ] is used to describe the priority of the microgrid. Each microgrid has its own binary variable as a trigger. When the trigger condition is met, the trigger is set to 1, indicating that the corresponding microgrid reserve is executed. The trigger conditions are: 1) The previous trigger is 1; if the current trigger has the highest priority, this condition can be ignored; 2) After the previous power supply microgrid is executed, the disadvantage microgrid still needs power support. The mathematical expression is as follows: Where: f∈F is the set of all microgrids; h∈H f A collection of microgrids that can provide power assistance. and Microgrid MG h The remaining dispatchable battery storage (SB) and gas turbine (MT) capacity, X f,t and Y f,t Yes H f ×1 matrix; x h,t and h,t It controls MG h The trigger for gas turbine startup, α h,t , β h,t They are The maximum allowable remaining dispatchable capacity, α is the limiting parameter, For MG h The gas turbine power generation capacity, For MG h The transition charge state of SB.

9. The microgrid energy management method based on hierarchical genetic algorithm according to claim 2 is characterized in that: The dual-objective genetic algorithm in step S15 comprises the following steps: 1) Initialize parameters, set chromosomes, population size, and the number of iterations to 1, and read data; 2) Randomly create individuals and check whether they meet the feasibility requirements. Keep the individuals that meet the feasibility requirements until the number reaches the set population size. 3) Calculate the fitness value according to the upper layer objective function; 4) Select individuals with higher fitness in the current population; 5) Cross the genes in the individuals, check the feasibility of the generated individuals, eliminate the individuals that lead to infeasible solutions, and continue crossover and mutation until the next generation of population is formed; 6) If the convergence condition is met, the optimal solution is returned and the algorithm ends; otherwise, go to 3); 7) The energy management system executes energy management and scheduling strategies according to the optimal solution.

10. A microgrid energy management method based on hierarchical genetic algorithm according to claim 9, characterized in that: In 2), the step of checking whether the individual satisfies feasibility is as follows: substitute each exchange variable of the upper-level individual into the lower-level individual, and calculate the lower-level model using the improved hierarchical genetic algorithm to check whether the lower-level optimization problem has a solution. If so, the feasibility is satisfied, otherwise, the feasibility is not satisfied.

Images