A capacity configuration method for integrated energy systems in desert areas

Abstract

The present invention discloses a capacity configuration method for an integrated energy system for desert areas. First, the CNN‑LSTM‑Attention model is used to predict the output of wind power and photovoltaic power and the user load power. Then, in view of the randomness and volatility of the output power of power sources such as wind power generation and photovoltaic power generation, an integrated energy storage device for hydrogen production, storage and power generation is constructed to alleviate the power fluctuation of the microgrid. After that, the economic models of the various components of the integrated energy system are constructed respectively, and then the profit objective functions of the various components of the integrated energy system are established. Finally, the Pied Kingfisher Optimization Algorithm (PKO) is used to iteratively search for the optimal solution with the maximization of profit as the optimization goal. The method of the present invention can effectively reduce the local abandonment of wind and solar energy while fully meeting the electricity demand of a given area, thereby realizing the effective utilization of resources.

Description

Technical Field

[0001] The invention belongs to the research field of integrated energy systems, and in particular relates to a capacity configuration method of an integrated energy system for desert areas. Background Art

[0002] At present, the main problems faced in resource development in areas rich in wind and solar resources such as deserts are that wind power generation and photovoltaic power generation are easily affected by the environment and climate, are highly uncontrollable, and waste resources. Therefore, when building wind and photovoltaic bases, energy storage systems are often required to suppress power fluctuations.

[0003] In order to improve the effective utilization of wind and solar resources in desert areas, it is necessary to design a comprehensive energy system capacity configuration method specifically for areas rich in wind and solar resources such as deserts, Gobi, and wastelands, so that it can fully meet the electricity demand of a given area while minimizing the local abandonment of wind and solar energy, and ultimately form an effective system for the comprehensive utilization and complementarity of multiple energy sources and a benign new energy utilization ecosystem in desert areas. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a capacity configuration method for an integrated energy system for desert areas.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A method for configuring capacity of an integrated energy system for desert areas comprises the following steps:

[0007] Step S1: Acquire meteorological data, historical power generation data of wind turbines, historical power generation data of photovoltaic devices, and user load data. Meteorological data includes wind speed, light intensity, temperature, and humidity; and perform data cleaning processing on the data;

[0008] Step S2: Establish CNN-LSTM-Attention model:

[0009] The CNN-LSTM-Attention model includes a CNN layer, an LSTM layer, an Attention layer, and a fully connected layer connected in sequence;

[0010] The CNN layer is a convolutional neural network, which is used to extract feature sequences and perform data dimensionality reduction on the feature sequences to obtain reduced-dimensional feature sequences. The LSTM layer is a long short-term memory neural network trained using reduced-dimensional feature sequences. Its output is calculated by the Attention layer using the attention mechanism and then input into the fully connected layer. The fully connected layer outputs the predicted power generation of the wind turbine, the predicted power generation of the photovoltaic device, and the predicted power of the user load.

[0011] Step S3: Establishing an integrated energy system capacity optimization configuration model:

[0012] Where: ω is the profit function of the integrated energy system; , , , , , , They are the electricity sales revenue, government subsidies, electricity purchase costs, wind and solar power abandonment penalty fees, load interruption compensation fees, annual average investment and installation costs, and annual average maintenance costs of the integrated energy system;

[0013] Average annual investment and installation cost for:

[0014]

[0015]

[0016]

[0017] Where: represents the average annual investment cost of the integrated hydrogen production-storage-power generation energy storage device, , , represent the average annual investment costs of fuel cells, hydrogen storage tanks, and electrolyzers, respectively; , , , , are the number of wind turbines, photovoltaic devices, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system; , , They are the installed capacities of wind turbines, photovoltaic devices, and fuel cells; is the capacity of the hydrogen storage tank; is the maximum amount of hydrogen that the electrolytic cell can produce; , , , , They are the unit costs of photovoltaic devices, wind turbines, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system; is the depreciation rate; is the service life; average annual maintenance cost for:

[0018]

[0019] Where: , , , , They are the maintenance costs per unit capacity of wind turbines, photovoltaic devices, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system;

[0020] Revenue from electricity sales in integrated energy systems The solution process:

[0021] First, calculate the comprehensive energy system in time Maximum power consumption :

[0022]

[0023] Where: is the time obtained in step S2 The predicted power of user load; is the maximum transmission power of the power transmission line; It's in time Maximum electrolytic power of a single electrolytic cell;

[0024] Then calculate the time Surplus power of integrated energy systems

[0025]

[0026]

[0027] Where: is the time obtained in step S2 Predicted power generation of wind turbines; is the time obtained in step S2 Predicted power generation from photovoltaic installations; For in time The electricity sales power of the integrated energy system;

[0028] Finally calculate the electricity sales revenue :

[0029]

[0030] Where: T is the scheduling period; is the electricity price; Δt is the time interval;

[0031] Electricity purchase costs for integrated energy systems The solution process:

[0032]

[0033]

[0034]

[0035] Where: For in time Calculated equivalent load power; For in time The amount of electricity purchased by the integrated energy system; The price of electricity purchased; It's in time Maximum power generation of a single fuel cell;

[0036] Penalty costs for wind and solar curtailment in integrated energy systems Model:

[0037]

[0038]

[0039] Where: Indicates at time The power generated by the integrated energy system is discarded; Indicates at time Penalty fees for wind and solar curtailment in integrated energy systems; is the penalty factor;

[0040] Load interruption compensation costs for integrated energy systems Model:

[0041]

[0042]

[0043] Where: For in time Total user load interruption power; For in time Interruption compensation paid by the integrated energy system; is the compensation factor;

[0044] Government subsidies for integrated energy systems Model:

[0045]

[0046] Where: Subsidized electricity prices for power generation from the integrated energy system given by the government;

[0047] Step S4: Use the Pied Kingfisher optimization algorithm to find the maximum value of the benefit function of the integrated energy system and obtain the optimal parameters for the capacity configuration of the integrated energy system.

[0048] Further, in step S4, the Kingfisher Optimization Algorithm is used to search for the optimal solution, which is composed of the following specific steps:

[0049] Step S4-1: Set initial parameters: population size, maximum number of iterations, dimension of decision variables, , , , , The lower bound of the decision variable corresponding to the minimum value of , , , , The upper bound of the decision variable corresponding to the maximum value of is set as the objective function to the opposite of the benefit function ω of the integrated energy system in step S3; generate the initial population:

[0050] In the formula: {X}_{i=}\left [ {{X}_{i1,}{X}_{i2,}{X}_{i3,}{X}_{i4,}{X}_{i5}} \right ] Indicates The position vector of each individual in the population; respectively

[0051] and , , , , correspond; is a random number between 0 and 1; N represents the population size; UB is the upper bound of the decision variable; LB is the lower bound of the decision variable;

[0052] Step S4-2: Fitness evaluation: For each individual in the population Calculate the initial fitness value:

[0053]

[0054] represents the fitness function of individual i; ω is the profit function of the integrated energy system in step S3; the optimal fitness of the initial population can be obtained and the corresponding optimal position vector ;

[0055] Step S4-3: Set the iteration loop counter b=1;

[0056] Step S4-4: For each individual in the population Do the following:

[0057] Randomly select a preset ratio The individuals in the population perform exploration operations, q is the first preset value, and the remaining individuals perform development operations;

[0058] The exploration operation consists of the following specific steps:

[0059] (1) Randomly select an individual that is different from the current individual ;

[0060] (2) Location update: When When it is less than q, the position update method is:

[0061]

[0062]

[0063] in, , Respectively represent The position of the individuals in the current iteration and the position of the next iteration of the population; randn represents the random number in the normal distribution; Dim is the dimension of the decision variable; represents the current position of the jth individual in the population; α represents the disturbance vector; B is the parameter variable; when p< <q, p is the second preset value, and the calculation process of parameter variable B is as follows:

[0064]

[0065]

[0066] Where: represents the maximum number of iterations; BF is a constant; exp() represents the natural exponential function; cos() represents the cosine function; π is the circumference of a circle; represents the disturbance variable; when When <p, the calculation process of parameter variable B is as follows:

[0067]

[0068]

[0069] In the formula: represents the fitness function of individual j; Indicates the slap rate;

[0070] The development operation for location update consists of the following specific steps:

[0071] when The position update process at q is as follows:

[0072]

[0073]

[0074]

[0075]

[0076] During the iteration process, represents the optimal position vector of the current iteration, represents the best fitness value in all iterations; HA represents hunting ability; is an auxiliary variable; is the calculation parameter; if─ω < Use Replace the optimal position vector ; If─ω > , then the optimal position vector constant;

[0077] when > The location update process is as follows:

[0078]

[0079]

[0080] In the formula: and are the position vectors of two individuals randomly selected from the population; The predation efficiency of Pied Kingfisher; Indicates taking the absolute value; and The fixed values ​​are respectively preset constants;

[0081] Step S4-5: Iteration loop counter b=b+1;

[0082] Step S4-6: Determine whether the current number of iterations has reached the maximum number of iterations. If so, output the optimal position vector As the optimal parameter for the capacity configuration of the integrated energy system, the optimal fitness function value The opposite number of is taken as the maximum value of the benefit function of the integrated energy system, otherwise, return to step S4-4.

[0083] The beneficial effects of the present invention are mainly manifested in that the method first uses the CNN-LSTM- Attention model to predict the power generation and load power of photovoltaic and wind power, and obtains the power curves of photovoltaic, wind power and load. Then, the present invention aims at the randomness and volatility of the output power of power sources such as wind power generation and photovoltaic power generation, and constructs an integrated energy storage device of hydrogen production, hydrogen storage and power generation to alleviate the power fluctuation of the microgrid. Then, a cost and benefit model of each part of the integrated energy system is established, the benefit function of each part of the integrated energy system is determined, and a capacity optimization configuration model of the integrated energy system is established. Finally, based on the data predicted by the CNN-LSTM- Attention model and the established capacity optimization configuration model of the integrated energy system, the spotted kingfisher optimization algorithm is used to find the optimal value. The experimental results show that the method of the present invention can effectively reduce the local abandonment of wind and solar energy while fully meeting the power demand of a given area, realize the effective utilization of resources, and improve the economy and reliability of the operation of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a flow chart of embodiment 1 of the present invention.

[0085] Figure 2 This is the energy architecture of the comprehensive energy system of the present invention.

[0086] Figure 3 This is a flow chart of the CNN-LSTM-Attention model of Example 1 of the present invention.

[0087] Figure 4This is a structural diagram of the CNN-LSTM-Attention model of Example 1 of the present invention.

[0088] Figure 5 This is a comparison chart of the prediction results of the CNN-LSTM-Attention model test set in Example 1 of the present invention.

[0089] Figure 6 : is the error histogram of the CNN-LSTM-Attention model of Example 1 of the present invention.

[0090] Figure 7 It is a comparison diagram of iterative convergence of the Pied Kingfisher optimization algorithm of Example 1 of the present invention.

[0091] Figure 8 This is a user load interruption diagram within a scheduling cycle of Example 1 of the present invention.

[0092] Fig. 9 This is a diagram of user wind and solar power abandonment within one scheduling cycle in Example 1 of the present invention.

[0093] Fig.10 This is a diagram of power changes of each power generation part and user load within one day in Example 1 of the present invention. DETAILED DESCRIPTION

[0094] Embodiment 1:

[0095] A method for configuring capacity of an integrated energy system for desert areas comprises the following steps:

[0096] S1: Obtain meteorological data, historical wind power generation data, historical photovoltaic power generation data, and user load data in desert areas, and perform data cleaning processing on them; meteorological data includes wind speed, light intensity, temperature, and humidity;

[0097] S2: Building a CNN-LSTM-Attention model:

[0098] The CNN-LSTM-Attention model includes a CNN layer, an LSTM layer, an Attention layer, and a fully connected layer connected in sequence;

[0099] The CNN layer is a convolutional neural network, which is used to extract feature sequences and perform data dimensionality reduction on the feature sequences to obtain reduced-dimensional feature sequences. The LSTM layer is a long short-term memory neural network trained using reduced-dimensional feature sequences. Its output is calculated by the Attention layer using the attention mechanism and then input into the fully connected layer to predict the power generation prediction data of a single wind turbine WT in a desert area. , photovoltaic array (PV) power generation forecast data , user load forecast data ;

[0100] Step S3: Establishing an integrated energy system capacity optimization configuration model:

[0101]

[0102] Where: ω is the profit function of the integrated energy system; , , , , , , They are the electricity sales revenue, government subsidies, electricity purchase costs, wind and solar power abandonment penalty fees, load interruption compensation fees, annual average investment and installation costs, and annual average maintenance costs of the integrated energy system;

[0103] Average annual investment and installation cost for:

[0104]

[0105]

[0106]

[0107] Where: represents the average annual investment cost of the integrated hydrogen production-storage-power generation energy storage device, , , represent the average annual investment costs of fuel cells, hydrogen storage tanks, and electrolyzers, respectively; , , , , are the number of wind turbines, photovoltaic devices, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system; , , They are the installed capacities of wind turbines, photovoltaic devices, and fuel cells; is the capacity of the hydrogen storage tank; is the maximum amount of hydrogen that the electrolytic cell can produce; , , , , are the unit costs of photovoltaic devices, wind turbines, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system respectively; m is the depreciation rate, which is 6%; r is the useful life, which is 15;

[0108] Average annual maintenance cost for:

[0109]

[0110] Where: , , , , The maintenance costs per unit capacity of wind turbines, photovoltaic devices, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system; the electricity sales revenue of the integrated energy system The solution process:

[0111] First calculate the comprehensive energy system time Maximum power consumption :

[0112]

[0113] Where: is the time obtained in step S2 The predicted power of user load; is the maximum transmission power of the power transmission line; It's in time Maximum electrolytic power of a single electrolytic cell;

[0114] Then calculate the time Surplus power of integrated energy systems :

[0115]

[0116]

[0117] Where: is the time obtained in step S2 Predicted power generation of wind turbines; is the time obtained in step S2 Predicted power generation from photovoltaic installations; For in time The electricity sales power of the integrated energy system;

[0118] Finally calculate the electricity sales revenue ;

[0119]

[0120] Where: T is the scheduling period, which we take as 480h here; is the electricity price; Δt is the time interval;

[0121] Electricity purchase costs for integrated energy systems The solution process:

[0122]

[0123]

[0124]

[0125] : For in time Equivalent load power; For in time The amount of electricity purchased by the integrated energy system; The price of electricity purchased; It's in time Maximum power generation of a single fuel cell;

[0126] Penalty costs for wind and solar curtailment in integrated energy systems Model:

[0127]

[0128]

[0129] Where: Indicates at time The power generated by the integrated energy system is discarded; Indicates at time Penalty fees for wind and solar curtailment in the integrated energy system; is the penalty factor, which is 0.3 yuan /

[0130] Load interruption compensation costs for integrated energy systems Model:

[0131]

[0132]

[0133] Where: For in time Total user load interruption power; For in time Interruption compensation payments made by the integrated energy system; The compensation factor is 0.9 yuan /

[0134] Government subsidies for integrated energy systems Model:

[0135]

[0136] Where: Subsidized electricity prices for power generation from the integrated energy system given by the government;

[0137] Step S4: Use the Pied Kingfisher optimization algorithm to find the maximum value of the benefit function of the integrated energy system and obtain the optimal parameters for the capacity configuration of the integrated energy system.

[0138] Further, in step S4, the Kingfisher Optimization Algorithm is used to search for the optimal solution, which is composed of the following specific steps:

[0139] Step S4-1: Set the initial parameters: the population size is 50, the maximum number of iterations is 1000, the dimension of the decision variable Dim is 5, and the lower bound of the decision variable LB is [1,10, 4, 1, 2] corresponding to S3 respectively. , , , , The minimum value of the five variables, the upper bound of the decision variable UB is [50,5000, 200, 40, 30] corresponding to S3 , , , , The maximum value of the five variables, the objective function is -ω, ω is the benefit function of the integrated energy system in S3; call the Initialization function to generate the initial population position and generate the initial population:

[0140]

[0141] In the formula: {X}_{i=}\left [ {{X}_{i1,}{X}_{i2,}{X}_{i3,}{X}_{i4,}{X}_{i5}} \right ] Indicates The position vector of each individual in the population; Respectively , , , , correspond; is a random number between 0 and 1; N represents the population size; UB is the upper bound of the decision variable; LB is the lower bound of the decision variable;

[0142] Step S4-2: Fitness evaluation: For each individual in the population Calculate the initial fitness value:

[0143]

[0144] represents the fitness function of individual i; ω is the profit function of the integrated energy system in step S3; the optimal fitness of the initial population can be obtained and the corresponding optimal position vector ;

[0145] Step S4-3: Set the iteration loop counter b=1;

[0146] Step S4-4: For each individual in the population Do the following:

[0147] Randomly select a preset ratio The individuals in the population perform exploration operations, q is the first preset value, and the remaining individuals perform development operations;

[0148] The exploration operation consists of the following specific steps:

[0149] (1) Randomly select an individual that is different from the current individual ;

[0150] (2) Location update: When When it is less than q, the position update method is:

[0151]

[0152]

[0153] in, , Respectively represent The position of the individuals in the current iteration and the position of the next iteration of the population; randn represents the random number in the normal distribution; Dim is the dimension of the decision variable; represents the current position of the jth individual in the population; α represents the disturbance vector; B is the parameter variable; when p< <q, p is the second preset value, and the calculation process of parameter variable B is as follows:

[0154]

[0155]

[0156] Where: Indicates the maximum number of iterations; is a constant; () represents the natural exponential function; () represents the cosine function; π is the circumference of a circle; represents the disturbance variable; when < When the parameter variable The calculation process is as follows:

[0157]

[0158]

[0159] In the formula: represents the fitness function of individual j; Indicates the slap rate;

[0160] The development operation for location update consists of the following specific steps:

[0161] when The position update process at q is as follows:

[0162]

[0163]

[0164]

[0165]

[0166] During the iteration process, represents the optimal position vector of the current iteration, represents the best fitness value in all iterations; HA represents hunting ability; is an auxiliary variable; is the calculation parameter; if─ω < Use Replace the optimal position vector ; If─ω > , then the optimal position vector constant;

[0167] when > The location update process is as follows:

[0168]

[0169]

[0170] In the formula: and are the position vectors of two individuals randomly selected from the population; The predation efficiency of Pied Kingfisher; Indicates taking the absolute value; and The fixed values ​​are respectively preset constants;

[0171] Step S4-5: Iteration loop counter b=b+1;

[0172] Step S4-6: Determine whether the current number of iterations has reached the maximum number of iterations. If so, output the optimal position vector As the optimal parameter for the capacity configuration of the integrated energy system, the optimal fitness function value The opposite number of is taken as the maximum value of the benefit function of the integrated energy system, otherwise, return to step S4-4.

[0173] In order to enable researchers in this field to better understand the process of capacity configuration of integrated energy systems, Figure 1 The flowchart is intuitive and easy to understand.

[0174] The present invention first uses the CNN-LSTM- Attention model to predict the power generation and load power of photovoltaic and wind power, and obtains the power curves of photovoltaic, wind power and load. Then, the present invention aims at the randomness and volatility of the output power of power sources such as wind power generation and photovoltaic power generation, and constructs an integrated energy storage device of hydrogen production, hydrogen storage and power generation to alleviate the power fluctuation of the microgrid. Then, the cost and benefit model of each part of the integrated energy system is established, the benefit function of each part of the integrated energy system is determined, and the capacity optimization configuration model of the integrated energy system is established. Finally, based on the data predicted by the CNN-LSTM-Attention model and the established capacity optimization configuration model of the integrated energy system, the spotted kingfisher optimization algorithm is used to find the optimal value. The experimental results show that the method of the present invention can effectively reduce the local abandonment of wind and solar energy while fully meeting the power demand of a given area, realize the effective utilization of resources, and improve the economy and reliability of the operation of the integrated energy system.

Claims

1. A method for configuring the capacity of an integrated energy system in desert areas, characterized in that: The following steps are involved: Step S1: Acquire meteorological data, historical power generation data of wind turbines, historical power generation data of photovoltaic devices, and user load data. Meteorological data includes wind speed, light intensity, temperature, and humidity; and perform data cleaning processing on the data; Step S2: Establish CNN-LSTM-Attention model: The CNN-LSTM-Attention model includes a CNN layer, an LSTM layer, an Attention layer, and a fully connected layer connected in sequence; The CNN layer is a convolutional neural network, which is used to extract feature sequences and perform data dimensionality reduction on the feature sequences to obtain reduced-dimensional feature sequences. The LSTM layer is a long short-term memory neural network trained using reduced-dimensional feature sequences. Its output is calculated by the Attention layer using the attention mechanism and then input into the fully connected layer. The fully connected layer outputs the predicted power generation of the wind turbine, the predicted power generation of the photovoltaic device, and the predicted power of the user load. Step S3: Establishing an integrated energy system capacity optimization configuration model: （1） Where: ω is the profit function of the integrated energy system; , , , , , , They are the electricity sales revenue, government subsidies, electricity purchase costs, wind and solar power abandonment penalty fees, load interruption compensation fees, annual average investment and installation costs, and annual average maintenance costs of the integrated energy system; Average annual investment and installation cost for: (2) (3) (4) Where: represents the average annual investment cost of the integrated hydrogen production-storage-power generation energy storage device, , , represent the average annual investment costs of fuel cells, hydrogen storage tanks, and electrolyzers, respectively; , , , , are the number of wind turbines, photovoltaic devices, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system; , , , are the installed capacities of wind turbines, photovoltaic devices, and fuel cells, respectively; is the capacity of the hydrogen storage tank; is the maximum amount of hydrogen that the electrolytic cell can produce; , , , , They are the unit costs of photovoltaic devices, wind turbines, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system; is the depreciation rate; is the useful life; Average annual maintenance cost for: (5) Where: , , , , They are the maintenance costs per unit capacity of wind turbines, photovoltaic devices, fuel cells, hydrogen storage tanks and electrolyzers in the integrated energy system; Revenue from electricity sales in integrated energy systems The solution process: First, calculate the comprehensive energy system in time Maximum power consumption : (6) Where: is the time obtained in step S2 is the time obtained in step S2 The predicted power of user load; The predicted power of user load; It's in time The maximum electrolytic power of a single electrolytic cell; Then calculate the time Surplus power of integrated energy systems : (7) (8) Where: is the time obtained in step S2 The predicted power generation of wind turbines; is the time obtained in step S2 The predicted power generation of the photovoltaic installation; For in time The electricity sales power of the integrated energy system; Finally calculate the electricity sales revenue : (9) Where: is the scheduling period; is the electricity price; Δt is the time interval; Electricity purchase costs for integrated energy systems The solution process: (10) (11) (12) Where: For in time Calculated equivalent load power; For in time The amount of electricity purchased by the integrated energy system; The price of electricity purchased; It's in time Maximum power generation of a single fuel cell; Penalty costs for wind and solar curtailment in integrated energy systems Model: (13) (14) Where: express Time integrated energy system discarded power generation; Indicates at time Penalty fees for wind and solar curtailment in the integrated energy system; is the penalty factor; Load interruption compensation costs for integrated energy systems Model: (15) (16) Where: For in time Total user load interruption power; For in time Interruption compensation payments made by the integrated energy system; is the compensation factor; Government subsidies for integrated energy systems Model: (17) Where: Subsidized electricity prices for power generation from the integrated energy system given by the government; Step S4: using the Kingfisher optimization algorithm to find the maximum value of the benefit function of the integrated energy system and obtain the optimal parameters for the capacity configuration of the integrated energy system; In step S4, the kingfisher optimization algorithm is used to search for the optimal solution, which is composed of the following specific steps: Step S4-1: Setting initial parameters: population size, maximum number of iterations, dimension of decision variables, , , , , The lower bound of the decision variable corresponding to the minimum value of , , , , The upper bound of the decision variable corresponding to the maximum value of is set as the objective function to the opposite of the benefit function ω of the integrated energy system in step S3; generate the initial population: （18） In the formula: {X}_{i}=\left [ {{X}_{i1,}{X}_{i2,}{X}_{i3,}{X}_{i4,}{X}_{i5}} \right ] Indicates The position vector of each individual in the population; , , , , Respectively , , , , correspond; A random number between 0 and 1; Indicates the population size; is the upper bound of the decision variable; is the lower bound of the decision variable; Step S4-2: Fitness evaluation: For each individual in the population Calculate the initial fitness value: (19) Represents an individual The fitness function of is the profit function of the integrated energy system in step S3; the optimal fitness of the initial population can be obtained and the corresponding optimal position vector ; Step S4-3: Setting the iteration loop counter ; Step S4-4: For each individual in the population Do the following: Randomly select a preset ratio The population individuals perform exploration operations, is the first preset value, and the remaining individuals perform development operations; The exploration operation consists of the following specific steps: (1) Randomly select an individual that is different from the current individual ; (2) Location update: When Less than When , the location update method is: (20) (21) in, , Respectively represent The position of the individuals in the population in the current iteration and the position of the individuals in the next iteration; represents a random number from a normal distribution; is the dimension of the decision variable; Indicates The current position of individuals in the population; α represents the disturbance vector; is a parameter variable; when < < hour, is the second preset value, parameter variable The calculation process is as follows: (22) (23) Where: Indicates the maximum number of iterations; is a constant; () represents the natural exponential function; () represents the cosine function; π is the circumference of a circle; represents the disturbance variable; when < When the parameter variable The calculation process is as follows: (24) (25) In the formula: Represents an individual The fitness function of Indicates the slap rate; The development operation for location update consists of the following specific steps: when > The location update process is as follows: (26) (27) (28) (29) During the iteration process, represents the optimal position vector of the current iteration, represents the best fitness value among all iterations; Indicates hunting ability; is an auxiliary variable; is the calculation parameter; if -ω < , then use Replace the optimal position vector ;If -ω ≥ , then the optimal position vector constant; when > The location update process is as follows: (30) (31) In the formula: and are the position vectors of two individuals randomly selected from the population; The predation efficiency of Pied Kingfisher; () means taking the absolute value; and The fixed values ​​are respectively preset constants; Step S4-5: Iterate the loop counter ; Step S4-6: Determine whether the current number of iterations has reached the maximum number of iterations. If so, output the optimal position vector As the optimal parameter for the capacity configuration of the integrated energy system, the optimal fitness function value The opposite number of is taken as the maximum value of the benefit function of the integrated energy system, otherwise, return to step S4-4.

Images