Micro-grid optimization scheduling method based on multi-target bee foraging algorithm

Abstract

The invention discloses a micro-grid optimization scheduling method based on a multi-target bee foraging algorithm. The method comprises the following steps: establishing a plurality of target functions for micro-grid optimization scheduling; determining multi-objective optimization scheduling related constraint conditions; parameters of the multi-target bee foraging algorithm are initialized; initializing reconnaissance bees according to the upper limit and the lower limit of each parameter of micro-grid optimization scheduling; initializing a nectar source storage library, and storing the Pareto optimal solution in the scout bees into the nectar source storage library; a selected nectar source is screened out from the nectar source storage library in a roulette mode; carrying out local search by the bees and updating the selected nectar source; carrying out local search by worker bees and updating the selected nectar source; the reconnaissance bees randomly search new nectar sources; and outputting a globally optimal solution and a fitness function value of the target function. According to the method, the Pareto non-dominated optimal solution set is utilized to improve the bee foraging algorithm, the multi-target optimization search efficiency and robustness are improved, and efficient solution of the micro-grid multi-target optimization scheduling parameters is realized.

Description

Technical Field

[0001] The invention belongs to the technical field of microgrid operation, and specifically is a microgrid optimization scheduling method based on a multi-objective bee foraging algorithm. Background Art

[0002] Microgrids are small power grids composed of distributed generation, loads, energy storage devices, and energy conversion devices. They can effectively improve the stability and power quality of a large number of distributed power sources connected to the main grid and realize the flexible application of distributed generation. However, the intermittent and unstable nature of distributed generation makes energy management more difficult. How to efficiently manage the energy of microgrids is a challenge faced by microgrid operation and scheduling.

[0003] At present, most of the technical solutions for microgrid optimization and scheduling only consider the economy and stability of microgrid operation, and pay less attention to pollutant emissions from traditional energy power generation. In addition, the microgrid optimization and scheduling process often needs to consider multiple optimization objectives, which may conflict with each other. How to fully consider the characteristics of different optimization objectives and solve problems such as slow convergence of optimization algorithms and easy to fall into local optimality has become an urgent need for microgrid optimization and scheduling. Summary of the invention

[0004] In view of the above problems, the present invention is proposed to provide a microgrid optimization scheduling method based on a multi-objective bee foraging algorithm to overcome the above problems or at least partially solve the above problems.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm comprises the following steps:

[0007] S1. Establish multiple objective functions for microgrid optimization scheduling;

[0008] S2. Determine the constraints related to multi-objective optimization scheduling;

[0009] S3, initializing and setting the parameters of the multi-objective bee foraging algorithm, the entire bee colony includes scout bees, foraging bees and worker bees;

[0010] S4, initializing the scout bee according to the upper and lower limits of each parameter of the microgrid optimization scheduling;

[0011] S5, initialize the nectar source storage, and store the Pareto optimal solution of the scout bees into the nectar source storage;

[0012] S6, selecting selected nectar sources from the nectar source storage using a roulette wheel method;

[0013] S7, honey bees conduct local search and update the selected nectar sources;

[0014] S8, waiting for the worker bees to conduct local search and update the selected nectar source;

[0015] S9, scout bees randomly search for new nectar sources;

[0016] S10. Output the global optimal solution of the objective function and its fitness function value.

[0017] Optionally, in step S1, the multiple objective functions of the microgrid optimization scheduling include an operation cost objective function expression, a pollutant emission objective function expression, and a nonlinear multi-objective optimization function;

[0018] The operating cost objective function expression is:

[0019]

[0020] Where, F1 represents the total cost of microgrid operation; T represents the optimal dispatching period; C WT (t) represents the operating cost of wind power in microgrid; C PV (t) represents the photovoltaic operation cost of the power grid; C FE (t) represents the operating cost of the microgrid fuel engine;

[0021] Among them, the wind power operation cost can be calculated according to the following formula:

[0022]

[0023] In the formula, C WT (t) represents the operating cost of wind power in microgrid; α wt represents the unit cost of wind power generation; P WT represents wind power generation; Δt represents time variation; T represents the optimization scheduling period;

[0024] The photovoltaic operation cost can be calculated according to the following formula:

[0025]

[0026] In the formula, C PV (t) represents the photovoltaic operation cost of the power grid; α pv Represents the unit cost of photovoltaic power generation; P PV represents photovoltaic power generation; Δt represents the time variation; T represents the optimization scheduling period;

[0027] The operating cost of a fuel engine can be calculated using the following formula:

[0028]

[0029] In the formula, CFE (t) represents the operating cost of the microgrid fuel engine; α fe represents the unit cost of fuel engine power generation; P FE represents the power generated by the fuel engine; Δt represents the time variation; T represents the optimization scheduling cycle;

[0030] The objective function expression of pollutant emissions is:

[0031]

[0032] In the formula, F2 represents the total amount of pollutant emissions; T represents the optimal scheduling period; β FE , γ FE , FE , η FE , FE represents the emission characteristic coefficient of the fuel generator; Δt represents the time change; P FE Indicates the power generated by the fuel engine;

[0033] The nonlinear multi-objective optimization function can be expressed as:

[0034] min[F1, F2]

[0035] Where F1 represents the total cost of microgrid operation; F2 represents the total amount of pollutant emissions.

[0036] Optionally, in step S2, the multi-objective optimization scheduling related constraints include:

[0037] Power balance constraints:

[0038] P WT +P PV +P FE =P LD +P LS

[0039] Where P WT Indicates wind power generation power; P PV Represents photovoltaic power generation; P FE Indicates the power generated by the fuel engine; P LD is the power load of the microgrid; P LS is the microgrid loss;

[0040] Microgrid output constraints:

[0041] P out_min ≤P out ≤P out_max

[0042] Where P out_max and P out_minThey represent the maximum output and minimum output of the microgrid respectively; Pout represents the actual output of the microgrid.

[0043] Optionally, in step S3, the parameters of the multi-objective bee foraging algorithm are initialized and set, and the parameters of the multi-objective bee foraging algorithm include the number of bee colonies, the algorithm stop condition, the neighborhood shrinkage coefficient, and the upper limit of the number of stagnation times of the bee search, wherein the upper limit of the number of stagnation times of the bee search is st It can be calculated according to the following formula:

[0044]

[0045] Where n c is the neighborhood shrinkage coefficient of the bee search; fix means rounding.

[0046] Optionally, the scout bee in step S4 uses the following search expression:

[0047]

[0048] In the formula, represents the position information of the i-th scout bee in the d-dimensional space; ub d and lb d They represent the upper and lower limits of the d-dimensional space respectively; rnd represents a random number between [0,1].

[0049] Optionally, in step S7, the steps of local search performed by bees are as follows:

[0050] S71, honey bees are evenly distributed to search near the selected nectar sources;

[0051] S72, for each selected nectar source, the honey bees assigned to the selected nectar source perform a random search near the nectar source;

[0052]

[0053] In the formula, Represents the position information of the i-th bee collector in the d-dimensional space; and They represent the upper and lower limits of the j-th selected honey source neighborhood search d-dimensional space respectively; rnd represents a random number between [0,1];

[0054] S73, finding the Pareto optimal solution searched by the honey bees, and comparing it with the Pareto optimal solution in the nectar source storage. If the Pareto optimal solution in the nectar source storage is updated, the number of search stagnation times of the current selected nectar source is reset, and the neighborhood range is restored to the original size;

[0055] S74, if the Pareto optimal solution in the nectar source storage is not updated, the search stagnation counter of the selected nectar source is increased by 1, and the search neighborhood of the selected nectar source is shrunk according to the following formula;

[0056] S nh (n+1)=n sc ·S nh (n) Among them, S nh (n) is the current search range; S nh (n+1) is the search range for the next iteration; n sc is the neighborhood shrinkage coefficient.

[0057] Optionally, in step S8, the steps for the worker bees to perform local search are as follows:

[0058] S81, evenly distributing the waiting worker bees to the selected nectar sources in the nectar source storage;

[0059] S82, the worker bee randomly selects a dimension from the selected nectar source where it is located, and performs the worker bee search update according to the following formula;

[0060]

[0061] Where d represents a randomly selected single dimension to be updated; Represents the position of the worker bee to be updated this time; and They represent the upper and lower limits of the j-th selected honey source neighborhood search d-dimensional space respectively; rnd represents a random number between [0,1];

[0062] S83, finding the Pareto optimal solution searched by the worker bees, and comparing it with the Pareto optimal solution in the nectar source storage library. If the Pareto optimal solution in the nectar source storage library is updated, the number of search stagnation times of the current selected nectar source is reset, and the neighborhood range is restored to the original size;

[0063] S84, if the Pareto optimal solution in the nectar source storage is not updated, the search stagnation counter of the selected nectar source is increased by 1;

[0064] S85. Determine whether the number of search stagnation times is greater than the upper limit of the number of stagnation times; if the number of search stagnation times is greater than the upper limit of the number of stagnation times, the corresponding selected nectar source will be abandoned in the global search stage.

[0065] Optionally, the scout bees account for 10% of the total number of the bee colony, and the number of honey bees and worker bees each account for 45% of the total number of the bee colony.

[0066] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0067] Aiming at the demand of microgrid optimization scheduling, the present invention proposes a microgrid optimization scheduling method based on a multi-objective bee foraging algorithm. The method takes the operation economy and pollution emission of the microgrid system as the main optimization objectives, constructs a multi-objective optimization function for microgrid optimization scheduling; improves the bee foraging algorithm by using the Pareto non-dominated optimal solution set, can more efficiently search for microgrid optimization scheduling parameters, improves the efficiency and robustness of multi-objective optimization search, and realizes efficient solution of microgrid multi-objective optimization scheduling parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 A flow chart of a microgrid optimization scheduling method based on a multi-objective bee foraging algorithm provided in an embodiment of the present application. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0070] See also Figure 1 , this embodiment provides a microgrid optimization scheduling method based on a multi-objective bee foraging algorithm, the method comprising the following steps:

[0071] S1. Establish multiple objective functions for microgrid optimization scheduling.

[0072] In step S1, the microgrid optimization scheduling multiple objective functions include an operation cost objective function expression, a pollutant emission objective function expression and a nonlinear multi-objective optimization function.

[0073] The operating cost objective function expression is:

[0074]

[0075] Where, F1 represents the total cost of microgrid operation; T represents the optimal dispatching period; C WT (t) represents the operating cost of wind power in microgrid; C PV (t) represents the photovoltaic operation cost of the power grid; C FE (t) represents the operating cost of the microgrid fuel engine.

[0076] Among them, the wind power operation cost can be calculated according to the following formula:

[0077]

[0078] In the formula, CWT (t) represents the operating cost of wind power in microgrid; α wt represents the unit cost of wind power generation; P WT represents wind power generation; Δt represents time variation; T represents the optimization scheduling period.

[0079] The photovoltaic operation cost can be calculated according to the following formula:

[0080]

[0081] In the formula, C PV (t) represents the photovoltaic operation cost of the power grid; α pv Represents the unit cost of photovoltaic power generation; P PV represents photovoltaic power generation; Δt represents the time change; T represents the optimization scheduling period.

[0082] The operating cost of a fuel engine can be calculated using the following formula:

[0083]

[0084] In the formula, C FE (t) represents the operating cost of the microgrid fuel engine; α fe represents the unit cost of fuel engine power generation; P FE represents the power generated by the fuel engine; Δt represents the time change; T represents the optimization scheduling period.

[0085] The objective function expression of pollutant emissions is:

[0086]

[0087] In the formula, F2 represents the total amount of pollutant emissions; T represents the optimal scheduling period; β FE , γ FE , FE , η FE , FE represents the emission characteristic coefficient of the fuel generator; Δt represents the time change; P FE Indicates the power generated by the fuel engine.

[0088] The nonlinear multi-objective optimization function can be expressed as:

[0089] min[F1, F2]

[0090] Where F1 represents the total cost of microgrid operation; F2 represents the total amount of pollutant emissions.

[0091] S2. Determine the constraints related to multi-objective optimization scheduling.

[0092] In step S2, the multi-objective optimization scheduling related constraints include:

[0093] Power balance constraints:

[0094] P WT +P PV +P FE =P LD +P LS

[0095] Where P WT Indicates wind power generation power; P PV Represents photovoltaic power generation; P FE Indicates the power generated by the fuel engine; P LD is the power load of the microgrid; P LS is the microgrid loss.

[0096] Microgrid output constraints:

[0097] P out_min ≤P out ≤P out_max Where P out_max and P out_min They represent the maximum output and minimum output of the microgrid respectively; Pout represents the actual output of the microgrid.

[0098] S3. Initialize and set the parameters of the multi-objective bee foraging algorithm. The entire bee colony includes scout bees, honey bees and worker bees. The scout bees account for 10% of the total number of the bee colony, and the number of honey bees and worker bees each account for 45% of the total number of the bee colony.

[0099] In step S3, the parameters of the multi-objective bee foraging algorithm are initialized and set. The parameters of the multi-objective bee foraging algorithm include the number of bee colonies, the algorithm stop condition, the neighborhood shrinkage coefficient, and the upper limit of the number of stagnation times of the bee search. st It can be calculated according to the following formula:

[0100]

[0101] Where n c is the neighborhood shrinkage coefficient of the bee search; fix means rounding.

[0102] S4. Initialize the scout bee according to the upper and lower limits of each parameter of the microgrid optimization scheduling.

[0103] The scout bee in step S4 uses the following search expression:

[0104]

[0105] In the formula, represents the position information of the i-th scout bee in the d-dimensional space; ub dand lb d They represent the upper and lower limits of the d-dimensional space respectively; rnd represents a random number between [0,1].

[0106] S5. Initialize the nectar source storage and store the Pareto optimal solution of the scout bees into the nectar source storage.

[0107] S6. Use the roulette method to select selected nectar sources from the nectar storage.

[0108] S7. Honey bees conduct local searches and update selected nectar sources.

[0109] In step S7, the local search steps performed by the honey bees are as follows:

[0110] S71. Honey bees, which account for 45% of the bee colony size, are evenly distributed to search near the selected nectar sources.

[0111] S72. For each selected nectar source, the honey bees assigned to the selected nectar source conduct a random search near the nectar source.

[0112]

[0113] In the formula, Represents the position information of the i-th bee collector in the d-dimensional space; and They represent the upper and lower limits of the d-dimensional space search for the j-th selected nectar source neighborhood; rnd represents a random number between [0,1].

[0114] S73. Find the Pareto optimal solution searched by the honey bees and compare it with the Pareto optimal solution in the nectar source storage library. If the Pareto optimal solution in the nectar source storage library is updated, the number of search stagnation times of the current selected nectar source is reset, and the neighborhood range is restored to its original size.

[0115] S74. If the Pareto optimal solution in the nectar source storage is not updated, the search stagnation counter of the selected nectar source is increased by 1, and the search neighborhood of the selected nectar source is shrunk according to the following formula.

[0116] S nh (n+1)=n sc ·S nh (n)

[0117] Among them, S nh (n) is the current search range; S nh (n+1) is the search range for the next iteration; n sc is the neighborhood shrinkage coefficient.

[0118] S8: Wait for the worker bees to conduct local search and update the selected nectar sources.

[0119] In step S8, the steps for the worker bees to perform local search are as follows:

[0120] S81. Evenly distribute the waiting bees to be workers to the selected nectar sources in the nectar storage.

[0121] S82. The worker bee randomly selects a dimension from the selected nectar source where it is located, and performs search and update according to the following formula.

[0122]

[0123] Where d represents a randomly selected single dimension to be updated; Represents the position of the worker bee to be updated this time; and They represent the upper and lower limits of the d-dimensional space search for the j-th selected nectar source neighborhood; rnd represents a random number between [0,1].

[0124] S83, find out the Pareto optimal solution searched by the worker bees, and compare it with the Pareto optimal solution in the nectar source storage library. If the Pareto optimal solution in the nectar source storage library is updated, the number of search stagnation times of the current selected nectar source is reset, and the neighborhood range is restored to the original size.

[0125] S84: If the Pareto optimal solution in the nectar source storage is not updated, the search stagnation counter of the selected nectar source is increased by 1.

[0126] S85. Determine whether the number of search stagnation times is greater than the upper limit of the number of stagnation times; if the number of search stagnation times is greater than the upper limit of the number of stagnation times, the corresponding selected nectar source will be abandoned in the global search stage.

[0127] S9. Scout bees randomly search for new nectar sources.

[0128] S10. Output the global optimal solution of the objective function and its fitness function value.

[0129] This embodiment takes the economic performance and pollution emissions of the microgrid system as the main optimization objectives, and constructs a multi-objective optimization function for microgrid optimization scheduling. In order to more efficiently search for microgrid optimization scheduling parameters, the Pareto non-dominated optimal solution set is used to improve the bee foraging algorithm, thereby improving the efficiency and robustness of the multi-objective optimization search and achieving efficient solution of the multi-objective optimization scheduling parameters of the microgrid.

[0130] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.

Claims

1. A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm, characterized in that: The method comprises the following steps: S1. Establish multiple objective functions for microgrid optimization scheduling; S2. Determine the constraints related to multi-objective optimization scheduling; S3, initializing and setting the parameters of the multi-objective bee foraging algorithm, the entire bee colony includes scout bees, foraging bees and worker bees; S4, initializing the scout bee according to the upper and lower limits of each parameter of the microgrid optimization scheduling; S5, initialize the nectar source storage, and store the Pareto optimal solution of the scout bees into the nectar source storage; S6, selecting selected nectar sources from the nectar source storage using a roulette wheel method; S7, honey bees conduct local search and update the selected nectar sources; S8, waiting for the worker bees to conduct local search and update the selected nectar source; S9, scout bees randomly search for new nectar sources; S10. Output the global optimal solution of the objective function and its fitness function value.

2. A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm as claimed in claim 1, characterized in that: In step S1, the microgrid optimization scheduling multiple objective functions include an operation cost objective function expression, a pollutant emission objective function expression and a nonlinear multi-objective optimization function; The operating cost objective function expression is: Where, F1 represents the total cost of microgrid operation; T represents the optimal dispatching period; C WT (t) represents the operating cost of wind power in microgrid; C PV (t) represents the photovoltaic operation cost of the power grid; C FE (t) represents the operating cost of the microgrid fuel engine; Among them, the wind power operation cost can be calculated according to the following formula: In the formula, C WT (t) represents the operating cost of wind power in microgrid; α wt represents the unit cost of wind power generation; P WT represents wind power generation; Δt represents time variation; T represents the optimization scheduling period; The photovoltaic operation cost can be calculated according to the following formula: In the formula, C PV (t) represents the photovoltaic operation cost of the power grid; α pv Represents the unit cost of photovoltaic power generation; P PV represents photovoltaic power generation; Δt represents the time variation; T represents the optimization scheduling period; The operating cost of a fuel engine can be calculated using the following formula: In the formula, C FE (t) represents the operating cost of the microgrid fuel engine; α fe represents the unit cost of fuel engine power generation; P FE represents the power generated by the fuel engine; Δt represents the time variation; T represents the optimization scheduling cycle; The objective function expression of pollutant emissions is: In the formula, F2 represents the total amount of pollutant emissions; T represents the optimal scheduling period; β FE , γ FE , FE , η FE , FE represents the emission characteristic coefficient of the fuel generator; Δt represents the time change; P FE Indicates the power generated by the fuel engine; The nonlinear multi-objective optimization function can be expressed as: min[F1,F2] Where F1 represents the total cost of microgrid operation; F2 represents the total amount of pollutant emissions.

3. A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm as claimed in claim 1, characterized in that: In step S2, the multi-objective optimization scheduling related constraints include: Power balance constraints: P WT +P PV +P FE =P LD +P LS Where P WT Indicates wind power generation power; P PV Represents photovoltaic power generation; P FE Indicates the power generated by the fuel engine; P LD is the power load of the microgrid; P LS is the microgrid loss; Microgrid output constraints: P out_min ≤P out ≤P out_max Where P out_max and P out_min They represent the maximum output and minimum output of the microgrid respectively; Pout represents the actual output of the microgrid.

4. A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm as claimed in claim 1, characterized in that: In step S3, the parameters of the multi-objective bee foraging algorithm are initialized and set. The parameters of the multi-objective bee foraging algorithm include the number of bee colonies, the algorithm stop condition, the neighborhood shrinkage coefficient, and the upper limit of the number of stagnation times of the bee search. st It can be calculated according to the following formula: Where n c is the neighborhood shrinkage coefficient of the bee search; fix means rounding.

5. A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm as claimed in claim 1, characterized in that: The scout bee in step S4 uses the following search expression: In the formula, represents the position information of the i-th scout bee in the d-dimensional space; ub d and lb d They represent the upper and lower limits of the d-dimensional space respectively; rnd represents a random number between [0,1].

6. A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm as claimed in claim 1, characterized in that: In step S7, the local search steps performed by the honey bees are as follows: S71, honey bees are evenly distributed to search near the selected nectar sources; S72, for each selected nectar source, the honey bees assigned to the selected nectar source perform a random search near the nectar source; In the formula, Represents the position information of the i-th bee collector in the d-dimensional space; and They represent the upper and lower limits of the d-dimensional space searched by the j-th selected nectar source neighborhood respectively; rnd represents a random number between [0,1]; S73, finding the Pareto optimal solution searched by the honey bees, and comparing it with the Pareto optimal solution in the nectar source storage. If the Pareto optimal solution in the nectar source storage is updated, the number of search stagnation times of the current selected nectar source is reset, and the neighborhood range is restored to the original size; S74, if the Pareto optimal solution in the nectar source storage is not updated, the search stagnation counter of the selected nectar source is increased by 1, and the search neighborhood of the selected nectar source is shrunk according to the following formula; S nh (n+1)=n sc ·S nh (n) Among them, S nh (n) is the current search range; S nh (n+1) is the search range for the next iteration; n sc is the neighborhood shrinkage coefficient.

7. A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm as claimed in claim 1, characterized in that: In step S8, the steps for the worker bees to perform local search are as follows: S81, evenly distributing the waiting worker bees to the selected nectar sources in the nectar source storage; S82, the worker bee randomly selects a dimension from the selected nectar source where it is located, and performs the worker bee search update according to the following formula; Where d represents a randomly selected single dimension to be updated; Represents the position of the worker bee to be updated this time; and They represent the upper and lower limits of the j-th selected honey source neighborhood search d-dimensional space respectively; rnd represents a random number between [0,1]; S83, finding the Pareto optimal solution searched by the worker bees, and comparing it with the Pareto optimal solution in the nectar source storage library. If the Pareto optimal solution in the nectar source storage library is updated, the number of search stagnation times of the current selected nectar source is reset, and the neighborhood range is restored to the original size; S84, if the Pareto optimal solution in the nectar source storage is not updated, the search stagnation counter of the selected nectar source is increased by 1; S85. Determine whether the number of search stagnation times is greater than the upper limit of the number of stagnation times; if the number of search stagnation times is greater than the upper limit of the number of stagnation times, the corresponding selected nectar source will be abandoned in the global search stage.

8. A microgrid optimization scheduling method based on a multi-objective bee foraging algorithm as claimed in claim 1, characterized in that: Scout bees account for 10% of the total number of bee colonies, while the number of honey bees and worker bees each account for 45% of the total number of bee colonies.

Images