Micro-grid optimization economic dispatching method based on improved flower pollination optimization algorithm

Abstract

The invention discloses a micro-grid optimization economic dispatching method based on an improved flower pollination optimization algorithm. The method comprises the following steps: firstly, obtaining the unit price, the operation coefficient, the maintenance coefficient, the processing coefficient and the charging and discharging efficiency of micro-grid equipment; secondly, an operation cost objective function and an environment cost objective function are established, the environment cost objective function comprises pollutant treatment cost and micro power source pollution treatment cost, and the total objective function is the sum of the pollutant treatment cost and the micro power source pollution treatment cost; meanwhile, various constraint conditions are adopted; solving by adopting an improved flower pollination algorithm, and generating a reverse solution through a generalized reverse learning mechanism to enhance population diversity; introducing a dynamic transition probability formula to adaptively balance global pollination and local pollination; and iterative optimization is carried out until an optimal scheduling scheme is output. According to the method, the convergence speed and optimization precision are remarkably improved, the total operation cost of the microgrid is effectively reduced, and the problem that a traditional algorithm is prone to local optimum is solved.

Description

Technical Field

[0001] This invention relates to the field of economic dispatch optimization technology for microgrids, and in particular to an optimized economic dispatch method for microgrids based on an improved flower pollination optimization algorithm. Background Technology

[0002] In recent years, with the continuous development of regional economy and society, energy consumption has become increasingly diversified, multi-sectoral, and multi-layered. To meet complex energy demands and promote the transformation of the energy structure, new energy microgrids have emerged. Microgrids are a globally popular new power system that can flexibly adjust the ratio of renewable and clean energy to load according to actual needs. Their operation can adapt to regionally differentiated energy consumption scenarios by flexibly allocating the ratio of various renewable and clean energy sources to load, meeting local power supply needs while simultaneously transmitting surplus electricity to other regions. With the accelerated large-scale application of renewable energy, improving its grid connection capacity and local consumption level has become a key development direction. The inherent intermittency and uncertainty of photovoltaic power generation pose significant challenges to the stable operation of microgrids. At the same time, microgrid clusters composed of multiple energy sources have complex coupling relationships between their internal subsystems. Against this backdrop, achieving effective and optimized scheduling of microgrids and their clusters is crucial for improving energy utilization efficiency and fully leveraging the complementary potential of multiple energy sources.

[0003] Domestic and international scholars have conducted extensive research on the problem of optimal economic dispatch of microgrids. Solving such problems using computer algorithms is currently a hot topic in microgrid economic optimization research. This is because optimal dispatch of microgrids requires comprehensive consideration of factors such as the supply capacity, cost, and energy storage capacity of various energy sources within the microgrid, involving multiple decision variables and constraints. However, its power generation is affected by weather, its structure is more complex than traditional power grids, and randomness and volatility are widespread in new energy output. Therefore, it is necessary to analyze different operating schemes and rationally plan the output schedule of micro-power sources to achieve optimal operation of the microgrid. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a microgrid optimization and economic dispatch method based on an improved flower pollination optimization algorithm. This method solves the problem that the structure of microgrids is very complex and there are difficulties in coordination and control. It has strong global search capabilities, introduces a generalized back learning mechanism to enhance the algorithm's ability to escape local optima, and has high accuracy in solving the microgrid optimization and economic dispatch problem, which can effectively reduce resource waste.

[0005] To achieve the above-mentioned technical effects, the technical solution adopted by the present invention is as follows: The microgrid optimization and economic dispatch method based on the improved flower pollination optimization algorithm includes the following steps: S1, obtain the unit price, operating coefficient, maintenance coefficient, processing coefficient and charging / discharging efficiency of microgrid equipment; S2, Establish the objective function for the operating cost of microgrid equipment. and environmental protection cost objective function ; S3 establishes the constraint conditions using power balance constraints, distributed power output constraints, and power interaction constraints. S4, Initialization parameters: including population size N, maximum number of iterations T, and optimization problem dimension Dim, and randomizing flower positions; S5 employs an opposing learning mechanism to compute the generalized inverse solution and calculates the values ​​of the original solution and the inverse solution; S6. Compare the values ​​of the original solution and the reverse solution: if the original solution is greater than the reverse solution, keep the population position unchanged; otherwise, use the reverse solution to replace the original solution to update the population position. S7, randomly generates dynamic probability k; S8: Set the dynamic probability p∈[0.2,0.9] according to different solutions; calculate the value of the new probability k using the formula. If k≤p, proceed to S9; if k>p, proceed to S10. S9, the algorithm performs a global pollination operation; S10, the algorithm performs local pollination operation; S11, calculate the fitness value of the new individual's position, and update the individual's optimal position and the global optimal position; S12: Determine if the maximum number of iterations of the algorithm has been reached. If yes, output the global optimal position; otherwise, increment the iteration count by 1 and return to S7.

[0006] Preferably, the microgrid optimized economic dispatch model includes constructing an objective function. : ; In the formula Operating costs; T For scheduling time; For consumption costs; For maintenance costs; Let t be the depreciation cost at time t; Compensation for power outage; This is for electricity trading fees.

[0007] Preferably, the microgrid optimized economic dispatch model includes establishing environmental governance costs and the cost of treating pollutants generated by micro-power sources during power generation as objective functions. : ; ; In the formula, For the cost of pollutant treatment, This is the cost coefficient. For operating costs, For pollutant emissions, For interactive power.

[0008] Preferably, the microgrid optimized economic dispatch model includes establishing a microgrid dispatch model with the objective function F being the sum of total operating cost and environmental cost: ; In the formula, For total cost, For operating costs, For environmental costs.

[0009] Preferably, the constraints described in step S3 include power balance constraints, distributed power output constraints, electrical energy interaction power constraints, battery state of charge constraints, diesel generator output power constraints, gas turbine output power constraints, photovoltaic generator output power constraints, and wind turbine output power constraints.

[0010] Preferably, the constraints are as follows: Establish power balance constraints: ; In the formula, This represents the active power output by the nth distributed generation unit at time point t. This refers to the net active power exchanged between the microgrid and the main grid. for t Microgrid load.

[0011] Establish distributed power output constraints: ; In the formula, , For the first n The upper and lower limits of the output of a distributed power source.

[0012] Establish power constraints for energy interaction: ; In the formula, , This represents the upper and lower limits of the power interaction between the microgrid and the main grid.

[0013] The state of charge (Soc) of a battery at any given time should be within its upper and lower limits:

[0014] In the formula, , This sets the upper and lower limits of the battery's output.

[0015] Establish diesel generator output power constraints:

[0016] In the formula, , These represent the upper and lower limits of the diesel generator's power.

[0017] S307, Establish gas turbine output power constraints:

[0018] In the formula, , These represent the upper and lower limits of gas turbine power.

[0019] S308, Establish photovoltaic generator output power constraints:

[0020] In the formula, , These represent the upper and lower limits of photovoltaic generator power.

[0021] S309, Establish wind turbine output power constraints:

[0022] In the formula, , These represent the upper and lower limits of wind turbine power. Preferably, in step S5, a generalized backward solution is calculated using a backward learning mechanism, and the values ​​of the original solution and the backward solution are calculated. The mathematical expression for the generalized backward learning mechanism is: ; The original solution in the formula , , and They are variables The lower and upper bounds, and the inverse solution vector are .

[0023] Preferably, in step S8, the formula for calculating the dynamic probability p is: ; in This represents the current iteration number. The maximum number of iterations, and These are parameters The minimum and maximum values; If the value is large, a global search is performed; as the iterations deepen further... The value is gradually decreased to perform a local search, and the search is refined to find the optimal solution.

[0024] Preferably, in step S9, the formula for performing the global pollination operation is: ; in, Indicates the first The flower in the Position at the next iteration Indicates the first The flower in the The position at the next iteration is a solution in the space. This represents the currently found globally optimal solution. It is a random number that follows a Lévy distribution.

[0025] Preferably, in step S10, the formula for performing the local pollination operation is: ; in, and Is with The positions of the other two flowers in the same neighborhood, It is a random number between [0,1].

[0026] The beneficial effects of this invention are as follows: 1. This invention uses the total operating cost of the microgrid as the objective function when establishing the microgrid economic dispatch model. During the dynamic dispatch of the microgrid, clean energy sources such as photovoltaic cells and wind turbines are prioritized. Renewable energy generation units have the highest priority as loads because they are environmentally friendly and have significant economic advantages; the cost only includes marginal cost operation and maintenance expenses, with no environmental remediation costs. When renewable energy output is limited and cannot meet demand, the system follows a tiered response mechanism: in the first stage, energy storage units are dispatched for power compensation; in the second stage, gas turbines are started and coordinated with the main grid for power procurement, dynamically adjusting output according to real-time electricity prices to ensure that the lowest-cost output is prioritized, achieving the goal of minimizing energy supply costs while ensuring power balance constraints. For situations of clean energy surplus, the system implements an energy consumption priority strategy: priority is given to charging energy storage devices, and surplus electricity is traded to other grids with limited output to achieve both power constraints and economic benefits. Furthermore, to achieve the goal of minimizing total cost, an algorithm is used to find the optimal operating scheme to solve the collaborative optimization problem of power interaction between gas turbines, energy storage units, and the network.

[0027] 2. This invention addresses the problems of low convergence accuracy and susceptibility to local optima in the whale optimization algorithm by employing a flower pollination optimization algorithm based on adaptive mechanism for dynamically adjusting the transition probability and opposing complement learning. Population aggregation is avoided by covering both ends of the solution space with the initial solution and the complementary solution. The complementary solution may be located in the region of the global optimum, significantly increasing the probability of discovering the global optimum. The diversity of the initial population makes the algorithm less susceptible to local optima trapping. The transition probability p∈[0.2,0.9], with p taking a value of approximately 0.9 in the early iterations, emphasizing global exploration to fully explore the global optimum and reduce the risk of local optima. In later iterations, p is approximately 0.2, emphasizing local development to increase the proportion of local development and accelerate convergence. The adaptive p ensures that: early global exploration discovers potential optimal regions; later local development refines the search for the optimal solution, while retaining some global exploration to avoid complete loss of diversity. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the optimization model solution process of the present invention; Figure 2 This is a schematic diagram of wind power, photovoltaic power and load power in an embodiment of the present invention. Detailed Implementation

[0029] Example 1: like Figure 1 As shown, the microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm includes the following steps: S1, obtain the unit price, operating coefficient, maintenance coefficient, processing coefficient and charging / discharging efficiency of microgrid equipment; S2, Establish the objective function for the operating cost of microgrid equipment. and environmental protection cost objective function ; S3 establishes the constraint conditions using power balance constraints, distributed power output constraints, and power interaction constraints. S4, Initialization parameters: including population size N, maximum number of iterations T, and optimization problem dimension Dim, and randomizing flower positions; S5 employs an opposing learning mechanism to compute the generalized inverse solution and calculates the values ​​of the original solution and the inverse solution; S6. Compare the values ​​of the original solution and the reverse solution: if the original solution is greater than the reverse solution, keep the population position unchanged; otherwise, use the reverse solution to replace the original solution to update the population position. S7, randomly generates dynamic probability k; S8: Set the dynamic probability p∈[0.2,0.9] according to different solutions; calculate the value of the new probability k using the formula. If k≤p, proceed to S9; if k>p, proceed to S10. S9, the algorithm performs a global pollination operation; S10, the algorithm performs local pollination operation; S11, calculate the fitness value of the new individual's position, and update the individual's optimal position and the global optimal position; S12: Determine if the maximum number of iterations of the algorithm has been reached. If yes, output the global optimal position; otherwise, increment the iteration count by 1 and return to S7.

[0030] Preferably, the microgrid optimized economic dispatch model includes constructing an objective function. : ; In the formula Operating costs; T For scheduling time; For consumption costs; For maintenance costs; Let t be the depreciation cost at time t; Compensation for power outage; This is for electricity trading fees.

[0031] Preferably, the microgrid optimized economic dispatch model includes establishing environmental governance costs and the cost of treating pollutants generated by micro-power sources during power generation as objective functions. : ; ; In the formula, For the cost of pollutant treatment, This is the cost coefficient. For operating costs, For pollutant emissions, For interactive power.

[0032] Preferably, the microgrid optimized economic dispatch model includes establishing a microgrid dispatch model with the objective function F being the sum of total operating cost and environmental cost: ; In the formula, For total cost, For operating costs, For environmental costs.

[0033] Preferably, the constraints described in step S3 include power balance constraints, distributed power output constraints, electrical energy interaction power constraints, battery state of charge constraints, diesel generator output power constraints, gas turbine output power constraints, photovoltaic generator output power constraints, and wind turbine output power constraints.

[0034] Preferably, the constraints are as follows: Establish power balance constraints: ; In the formula, This represents the active power output by the nth distributed generation unit at time point t. This refers to the net active power exchanged between the microgrid and the main grid. for t Microgrid load.

[0035] Establish distributed power output constraints: ; In the formula, , For the first n The upper and lower limits of the output of a distributed power source.

[0036] Establish power constraints for energy interaction: ; In the formula, , This represents the upper and lower limits of the power interaction between the microgrid and the main grid.

[0037] The state of charge (Soc) of a battery at any given time should be within its upper and lower limits:

[0038] In the formula, , This sets the upper and lower limits of the battery's output.

[0039] Establish diesel generator output power constraints:

[0040] In the formula, , These represent the upper and lower limits of the diesel generator's power.

[0041] S307, Establish gas turbine output power constraints:

[0042] In the formula, , These represent the upper and lower limits of gas turbine power.

[0043] S308, Establish photovoltaic generator output power constraints:

[0044] In the formula, , These represent the upper and lower limits of photovoltaic generator power.

[0045] S309, Establish wind turbine output power constraints:

[0046] In the formula, , These represent the upper and lower limits of wind turbine power. Preferably, in step S5, a generalized backward solution is calculated using a backward learning mechanism, and the values ​​of the original solution and the backward solution are calculated. The mathematical expression for the generalized backward learning mechanism is: ; The original solution in the formula , , and They are variables The lower and upper bounds, and the inverse solution vector are .

[0047] Preferably, in step S8, the formula for calculating the dynamic probability p is: ; in This represents the current iteration number. The maximum number of iterations, and These are parameters The minimum and maximum values; If the value is large, a global search is performed; as the iterations deepen further... The value is gradually decreased to perform a local search, and the search is refined to find the optimal solution.

[0048] Preferably, in step S9, the formula for performing the global pollination operation is: ; in, Indicates the first The flower in the Position at the next iteration Indicates the first The flower in the The position at the next iteration is a solution in the space. This represents the currently found globally optimal solution. It is a random number that follows a Lévy distribution.

[0049] Preferably, in step S10, the formula for performing the local pollination operation is: ; in, and Is with The positions of the other two flowers in the same neighborhood, It is a random number between [0,1].

[0050] Example 2: Select typical day's wind power, solar power, and load forecast power as follows: Figure 2 As shown, Table 1 lists the parameter information of each distributed power source in the microgrid, Table 2 gives the pollution emission coefficient of each type of distributed power source, Table 3 shows the energy storage system data, and Table 4 includes the purchase and sale electricity price information for each time period.

[0051] Table 1: Parameters and Operating Costs;

[0052] Table 2: Emission coefficients and costs;

[0053] Table 3: Battery energy storage parameters;

[0054] Table 4: Time-of-use electricity pricing;

[0055] The optimized total cost data is shown in Table 5: Table 5: Algorithm Comparison and Analysis;

[0056] From the above Figure 2 As shown in Table 5, existing whale optimization algorithms suffer from low convergence accuracy and are prone to getting trapped in local optima. This method employs a flower pollination optimization algorithm based on adaptive mechanisms to dynamically adjust the transition probability and anti-optimal learning. By covering both ends of the solution space with the initial and opposing solutions, population aggregation is avoided. Opposing solutions may be located in the region of the global optimum, significantly increasing the probability of finding the global optimum. Initial population diversity makes the algorithm less susceptible to local optima trapping. The transition probability p∈[0.2,0.9], with p valued at approximately 0.9 in early iterations, emphasizing global exploration (long-distance pollination) to fully explore the global environment and reduce the risk of local optima. In later iterations, p is approximately 0.2, emphasizing local development (short-distance pollination) to increase the proportion of local development and accelerate convergence. The adaptive p ensures that: early global exploration discovers potential optimal regions; later local development refines the search for optimal solutions, while retaining 20% ​​of global exploration to avoid complete loss of diversity.

[0057] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The embodiments and features described in these embodiments can be arbitrarily combined without conflict. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A microgrid optimized economic dispatch method based on an improved flower pollination optimization algorithm, characterized in that, Includes the following steps: S1, obtain the unit price, operating coefficient, maintenance coefficient, processing coefficient and charging / discharging efficiency of microgrid equipment; S2, Establish the objective function for the operating cost of microgrid equipment. and environmental protection cost objective function ; S3 establishes the constraint conditions using power balance constraints, distributed power output constraints, and power interaction constraints. S4, Initialization parameters: including population size N, maximum number of iterations T, and optimization problem dimension Dim, and randomizing flower positions; S5 employs an opposing learning mechanism to compute the generalized inverse solution and calculates the values ​​of the original solution and the inverse solution; S6. Compare the values ​​of the original solution and the reverse solution: if the original solution is greater than the reverse solution, keep the population position unchanged; otherwise, use the reverse solution to replace the original solution to update the population position. S7, randomly generates dynamic probability k; S8, set the dynamic probability p∈[0.2,0.9] according to different solution objectives; The new probability k is calculated using the formula. If k ≤ p, proceed to S9; if k > p, proceed to S10. S9, the algorithm performs a global pollination operation; S10, the algorithm performs local pollination operation; S11, calculate the fitness value of the new individual's position, and update the individual's optimal position and the global optimal position; S12: Determine if the maximum number of iterations of the algorithm has been reached. If yes, output the global optimal position; otherwise, increment the iteration count by 1 and return to S7.

2. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 1, characterized in that, The microgrid optimized economic dispatch model includes constructing an objective function. : ； In the formula Operating costs; T For scheduling time; For consumption costs; For maintenance costs; Let t be the depreciation cost at time t; Compensation for power outage; This is for electricity trading fees.

3. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 2, characterized in that, The microgrid optimized economic dispatch model includes setting environmental governance costs and the cost of treating pollutants generated by micro-power sources during power generation as objective functions. : ； ； In the formula, For the cost of pollutant treatment, This is the cost coefficient. For operating costs, For pollutant emissions, For interactive power.

4. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 3, characterized in that, The microgrid optimized economic dispatch model includes establishing a microgrid dispatch model objective function F, which is the sum of total operating cost and environmental cost. ； In the formula, For total cost, For operating costs, For environmental costs.

5. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 1, characterized in that, The constraints described in step S3 include power balance constraints, distributed power output constraints, electrical energy interaction power constraints, battery state of charge constraints, diesel generator output power constraints, gas turbine output power constraints, photovoltaic generator output power constraints, and wind turbine output power constraints.

6. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 5, characterized in that, The specific constraints are as follows: Establish power balance constraints: ； In the formula, This represents the active power output of the nth distributed generation unit at time point t; This refers to the net active power exchanged between the microgrid and the main grid. for t Microgrid load; Establish distributed power output constraints: ； In the formula, , For the first n Upper and lower limits of distributed power generation output; Establish power constraints for energy interaction: ； In the formula, , This represents the upper and lower limits of the power interaction between the microgrid and the main grid. The state of charge (Soc) of a battery at any given time should be within its upper and lower limits: In the formula, , The upper and lower limits of battery output; Establish diesel generator output power constraints: In the formula, , These are the upper and lower limits of the diesel generator's power. S307, Establish gas turbine output power constraints: In the formula, , These are the upper and lower limits of gas turbine power; S308, Establish photovoltaic generator output power constraints: In the formula, , These are the upper and lower limits of photovoltaic generator power. S309, Establish wind turbine output power constraints: In the formula, , These represent the upper and lower limits of wind turbine power.

7. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 1, characterized in that, In step S5, a generalized backward solution is calculated using a backward learning mechanism, and the values ​​of the original solution and the backward solution are calculated. The mathematical expression for the generalized backward learning mechanism is: ； The original solution in the formula , , and They are variables The lower and upper bounds, and the inverse solution vector are .

8. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 1, characterized in that, In step S8, the formula for calculating the dynamic probability p is: ； in This represents the current iteration number. The maximum number of iterations, and These are parameters The minimum and maximum values; If the value is large, a global search is performed; as the iterations deepen further... The value is gradually decreased to perform a local search, and the search is refined to find the optimal solution.

9. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 1, characterized in that, In step S9, the formula for performing the global pollination operation is as follows: ； in, Indicates the first The flower in the Position at the next iteration Indicates the first The flower in the The position at the next iteration is a solution in the space. This represents the currently found globally optimal solution. It is a random number that follows a Lévy distribution.

10. The microgrid optimized economic dispatch method based on the improved flower pollination optimization algorithm according to claim 1, characterized in that, In step S10, the formula for performing local pollination is as follows: ； in, and Is with The positions of the other two flowers in the same neighborhood, It is a random number between [0,1].

Images