Electric heating interconnection virtual power plant economic dispatching solving method

Abstract

The invention relates to the technical field of economic dispatching, in particular to an electrothermal interconnected virtual power plant economic dispatching solving method, which comprises the following steps: acquiring information of equipment in a to-be-solved virtual power plant and time-of-use electricity price of the day, and determining a main optimization objective function according to the type of a solved unit; time decoupling is carried out according to the peak-valley electricity price, and electric thermal decoupling is carried out on the electric thermal interconnection equipment; the method comprises the following steps: establishing a distributed multi-agent system based on a Java agent development framework, and classifying different agents; improved transverse crossing and improved longitudinal crossing are respectively executed in the intelligent body; the intelligent agents are updated by adopting a similar cross interchange strategy; each distributed sub optimizer performs population updating by taking the global optimal particle as initialization data and stores a global optimal population and a corresponding global optimal fitness value; the number of optimization iterations is met, and virtual power plant economic dispatching is output to solve and optimize optimal population information. The method can reduce the communication burden of solving the economic dispatching problem of the virtual power plant, and reduces the solving difficulty of the problem.

Description

Technical Field

[0001] The present invention relates to the technical field of economic dispatch, and more specifically, to a method for solving the economic dispatch of an electric-heat interconnected virtual power plant. Background Art

[0002] As an important energy carrier, the virtual power plant provides a more sustainable and environmentally friendly solution for regional energy supply, and makes an important contribution to achieving the low-carbon goal.

[0003] The main goal of the virtual power plant is to integrate and allocate various distributed energy sources, and minimize the adverse effects brought by the volatility of uncontrollable distributed energy sources within the virtual power plant. Existing technologies mostly use mathematical programming methods such as robust optimization and stochastic optimization for optimal dispatch. However, the mathematical programming method mainly solves linear convex optimization problems. When the constraints increase, the nonlinearity of the problem to be solved is difficult to be linearized by the mathematical programming method.

[0004] The prior art discloses a decoupling and dimension reduction solution method for multi-region static economic dispatch, which obtains the information of multi-region static economic dispatch units and establishes a main optimization objective function; uses a cross-in-cross algorithm to decouple multi-region units to obtain N sub-populations and distributed sub-optimizers; based on the sub-populations, establishes a sub-association objective function between the sub-optimizer and the remaining sub-populations, and calculates the sub-objective functions of each sub-population; the populations of each sub-optimizer are respectively subjected to parallel horizontal crossover and vertical crossover; then calculates the neighborhood crossover new population through the neighborhood population crossover mechanism, and accelerates the update of the distributed sub-optimizer through the mutation mechanism; performs population update to obtain sub-global optimal particles; when the update iteration times are satisfied, outputs the optimal population information of the multi-region static economic dispatch solution optimization. In this solution, the swarm intelligence optimization algorithm based on the population concept can handle complex economic dispatch problems by optimizing the objective function, establishing a variable population, and updating the population. However, when the number of devices in the virtual power plant increases, the excessive communication volume will seriously reduce the calculation efficiency and is prone to information loss, resulting in the algorithm being unable to converge. Summary of the Invention

[0005] The purpose of the present invention is to overcome the deficiency of the large communication burden in the economic dispatch problem of large-scale virtual power plants in the prior art, and provide a method for solving the economic dispatch of an electric-heat interconnected virtual power plant, which reduces the communication burden of solving the economic dispatch problem of the virtual power plant and reduces the difficulty of solving the problem.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is:

[0007] Provide a method for solving the economic dispatch of an electric-heat interconnected virtual power plant, including the following steps:

[0008] S1: Obtain the information of the equipment in the virtual power plant to be solved and the time-of-use electricity price of the current day, and establish the main optimization objective function according to the type of unit to be solved;

[0009] S2: Decouple the main optimization objective function in step S1 according to the peak-valley electricity price to obtain the first sub-objective function;

[0010] S3: Decouple the electric-thermal interconnected equipment in the first sub-objective function in step S2 to obtain the second sub-objective functions reflecting the power grid and the heat network of the virtual power plant to be solved respectively;

[0011] S4: Based on the Java intelligent agent development framework, establish a distributed multi-agent system, use the improved horizontal and vertical crossover algorithm to solve the second sub-objective function in step S3 to obtain the sub-population and the distributed sub-optimizer, arrange the distributed sub-optimizer in the distributed multi-agent system, replace each distributed sub-optimizer with an intelligent agent, and classify different intelligent agents;

[0012] S5: The intelligent agents respectively execute the improved horizontal crossover and the improved vertical crossover, and use the improved horizontal and vertical crossover optimization algorithm to solve;

[0013] S6: The intelligent agents use the same-kind crossover and exchange strategy to update;

[0014] S7: Each distributed sub-optimizer uses the global optimal particle of the updated distributed sub-optimizer in step S6 as the initialization data for population update and stores the global optimal population and the corresponding global optimal fitness value;

[0015] S8: Meet the optimization iteration times and output the optimal population information of the economic dispatch solution of the virtual power plant.

[0016] For the economic dispatch solution method of the electric-thermal interconnected virtual power plant of the present invention, the main optimization objective function is established according to the type of unit to be solved, the time scale of the electric-thermal interconnected virtual power plant is divided based on the peak-valley electricity price. At the same time, the power transmission network and the heat transmission network of the virtual power plant are decentralized and autonomous by means of electric-thermal decoupling, so as to achieve the double decoupling of the electric-thermal interconnected virtual power plant. Then, the second sub-objective function is calculated based on the divided sub-problems. Through two-dimensional reduction, the defect of local optimality of large-scale variable optimization problems is avoided, and the convergence accuracy is improved. The sub-objective multi-population interaction operation co-evolves, and the improved horizontal and vertical crossover optimization algorithm and the same-kind intelligent agent crossover and exchange strategy are used to solve the sub-problems, accelerating the convergence speed and improving the ability to mine detailed information of the convergence algorithm. Finally, when the iteration times are met, the optimal population information of the economic dispatch solution of the virtual power plant is output. In the present invention, the heat and power grids only need to exchange a small amount of boundary variables and the virtual target factors are globally adjusted respectively to obtain the multi-objective solution, reducing the communication burden and the difficulty of solving the problem.

[0017] Preferably, in step S1, the main optimization objective function is:

[0018]

[0019] In the formula, represents the electricity sales volume of the virtual power plant to the system at the t-th moment; represents the electricity sales price at the t-th moment; represents the electricity purchase volume of the virtual power plant from the system at the t-th moment; represents the electricity purchase price at the t-th moment; pen t represents the penalty coefficient at the t-th moment; represents the fuel cost at the t-th moment; represents the operation and maintenance cost at the t-th moment; N t represents the total time period, t ∈ [1, 2, ……, N t .

[0020] Preferably, in step S2, according to the peak, valley, and normal time periods corresponding to the time-of-use electricity price, the main optimization objective function in step S1 is decoupled and divided into N a peak moments, N v valley moments, and N f normal moments, where N a , N v and N f are all less than N t and N a + N v + N f = N t ;

[0021] The first sub-objective function at peak time obtained after decoupling is:

[0022]

[0023] In the formula, represents the first sub-objective function at peak time after time decoupling; represents the electricity sales volume of the virtual power plant to the system at the t a -th moment; represents the electricity sales price at the t a -th moment; represents the electricity purchase volume of the virtual power plant from the system at the t a -th moment; represents the electricity purchase price at the t a -th moment; represents the penalty coefficient at the t a -th moment; represents the fuel cost at the t a -th moment; represents the t aOperation and maintenance cost at a certain moment; α p Represents the electricity coefficient of the combined heat and power unit; α h Represents the heat coefficient of the combined heat and power unit; Represents the t a Heat output of the combined heat and power unit at a certain moment; aff bl Represents the energy efficiency coefficient of the boiler; Represents the t a Heat output of the boiler at a certain moment; Q gas Represents the price of natural gas; Represents the t a Operation and maintenance cost of the energy storage device at a certain moment; Represents the t a Power generation of the combined heat and power unit at a certain moment; Z pchp Represents the operation and maintenance cost per unit of electricity generated by the combined heat and power unit; Z pchp Represents the operation and maintenance cost per unit of heat generated by the combined heat and power unit; Represents the t a Power generation of the photovoltaic at a certain moment; Z ppv Represents the operation and maintenance cost per unit of electricity generated by the photovoltaic; Represents the power generation of the wind turbine at time ta; Z pw Represents the operation and maintenance cost per unit of electricity generated by the wind turbine; Z bl Represents the operation and maintenance cost per unit of heat generated by the boiler;

[0024] The first sub-objective function at the valley time after decoupling is:

[0025]

[0026] In the formula, Represents the first sub-objective function at the valley time after time decoupling; Represents the t v Electricity sold by the virtual power plant to the system at a certain moment; Represents the t v Selling electricity price at a certain moment; Represents the t v Electricity purchased by the virtual power plant from the system at a certain moment; Represents the t v Purchasing electricity price at a certain moment; Represents the t v Penalty coefficient at a certain moment; Represents the t v Fuel cost at a certain moment; Represents the t v Operation and maintenance cost at a certain moment; Represents the t v Heat output of the combined heat and power unit at a certain moment; Represents the t vThe calorific value of the boiler at a certain moment; Indicates the t v The operation and maintenance cost of the energy storage device at a certain moment; Indicates the t v The power generation of the combined heat and power unit at a certain moment; Indicates the t v The power generation of the photovoltaic at a certain moment; Indicates the t v The power generation of the wind turbine at a certain moment;

[0027] The first sub-objective function in normal times obtained after decoupling is:

[0028]

[0029] In the formula, Indicates the first sub-objective function in normal times after time decoupling; Indicates the t f The electricity sales volume of the virtual power plant to the system at a certain moment; Indicates the t f The electricity sales price at a certain moment; Indicates the t f The electricity purchase volume of the virtual power plant from the system at a certain moment; Indicates the t f The electricity purchase price at a certain moment; Indicates the t f The penalty coefficient at a certain moment; Indicates the t f The fuel cost at a certain moment; Indicates the t f The operation and maintenance cost at a certain moment; Indicates the t f The calorific value of the combined heat and power unit at a certain moment; Indicates the t f The calorific value of the boiler at a certain moment; Indicates the t f The operation and maintenance cost of the energy storage device at a certain moment; Indicates the t f The power generation of the combined heat and power unit at a certain moment; Indicates the t f The power generation of the photovoltaic at a certain moment; Indicates the t f The power generation of the wind turbine at a certain moment.

[0030] Preferably, in step S3, the electro-thermal interconnection device in the first sub-objective function in step S2 is subjected to electro-thermal decoupling to obtain:

[0031] The second sub-objective function of the power grid during peak hours Is:

[0032]

[0033] The second sub-objective function of the peak-time heat network is:

[0034]

[0035] The second sub-objective function of the valley-time power grid is:

[0036]

[0037] The second sub-objective function of the valley-time heat network is:

[0038]

[0039] The second sub-objective function of the normal-time power grid is:

[0040]

[0041] The second sub-objective function of the normal-time heat network is:

[0042]

[0043] Preferably, in step S4, the sub-population corresponding to the second sub-objective function of the peak-time power grid expression is:

[0044]

[0045] The sub-population corresponding to the second sub-objective function of the peak-time heat network expression is:

[0046]

[0047] The sub-population corresponding to the second sub-objective function of the valley-time power grid expression is:

[0048]

[0049] The sub-population corresponding to the second sub-objective function of the valley-time heat network expression is:

[0050]

[0051] The sub-population corresponding to the second sub-objective function of the normal-time power grid expression is:

[0052]

[0053] The sub-population corresponding to the second sub-objective function of the normal heat power grid The expression is as follows:

[0054]

[0055] The population expression corresponding to the global total population X is:

[0056]

[0057] Preferably, in step S4, the agents are divided into two categories, where:

[0058] The expression of one type of agent is:

[0059]

[0060] In the formula; represents the sub-optimizer for solving the sub-objective function of the peak-time power grid; represents using the CSO optimization algorithm to solve the sub-objective function of the peak-time power grid; represents the sub-optimizer for solving the sub-objective function of the valley-time power grid; represents using the CSO optimization algorithm to solve the sub-objective function of the valley-time power grid; represents the sub-optimizer for solving the sub-objective function of the normal-time power grid; represents using the CSO optimization algorithm to solve the sub-objective function of the normal-time power grid;

[0061] The expression of the second type of agent is:

[0062]

[0063] In the formula; represents the sub-optimizer for solving the sub-objective function of the peak-time heat power grid; represents using the CSO optimization algorithm to solve the sub-objective function of the peak-time heat power grid; represents the sub-optimizer for solving the sub-objective function of the valley-time heat power grid; represents using the CSO optimization algorithm to solve the sub-objective function of the valley-time heat power grid; represents the sub-optimizer for solving the sub-objective function of the normal-time heat power grid; represents using the CSO optimization algorithm to solve the sub-objective function of the normal-time heat power grid.

[0064] Preferably, in step S5, a random differential mutation strategy is adopted inside each distributed sub-optimizer for horizontal crossover:

[0065]

[0066]

[0067] respectively represent the solutions after horizontal crossover of the agents corresponding to each second sub-objective function; j and k respectively represent the particles of the agent population.

[0068]

[0069] respectively represent the solutions randomly generated by the agents corresponding to each second sub-objective function. respectively represent all dimensions of the j-th particle of the corresponding sub-optimizer. respectively represent all dimensions of the k-th particle of the corresponding sub-optimizer; r and n represent random numbers uniformly distributed between [0, 1]; c represents a random number uniformly distributed between [-1, 1].

[0070] Preferably, in step S5, vertical crossover is performed inside each distributed sub-optimizer:

[0071]

[0072] In the formula, respectively represent the solutions after vertical crossover of the agents corresponding to each second sub-objective function; d1 and d2 respectively represent the d1-th dimension and the d2-th dimension, d1≠d2 and d2∈(1, D), where D represents the total dimension of the problem to be solved.

[0073] Preferably, in step S6, the expression of the same-class crossover and swap strategy is:

[0074]

[0075]

[0076] In the formula, respectively represent the solutions after the same-class crossover and swap of the corresponding sub-optimizer. respectively represent all dimensions of the z-th particle of the corresponding sub-optimizer. respectively represent all dimensions of the ν-th particle of the corresponding sub-optimizer. respectively represent all dimensions of the u-th particle of the corresponding sub-optimizer; r1 and r2 respectively represent random numbers uniformly distributed between [0, 1].

[0077] Preferably, the population update includes:

[0078] S701: Each distributed sub-optimizer calculates the second sub-objective function respectively, updates the population according to the greedy extreme value, and obtains the sub-global optimal particles respectively.

[0079] S702: Obtain the total global optimal particle.

[0080] S703: Calculate the global optimal fitness.

[0081] Compared with the prior art, the beneficial effects produced by the present invention are as follows:

[0082] Through time decoupling and electro-thermal decoupling, secondary dimensionality reduction is achieved, avoiding the local optimal defect of large-scale variable optimization problems and improving the convergence accuracy; an improved cross optimization algorithm and a cross-exchange strategy of similar agents are used to solve sub-problems, accelerating the convergence speed and improving the ability to mine detailed information of the convergence algorithm; the heat and power grids only need to exchange a small number of boundary variables and the virtual target factors are globally adjusted respectively to obtain multi-objective solutions, reducing the communication burden and the difficulty of problem solving. Description of the Drawings

[0083] Figure 1 It is a flowchart of the economic dispatch solution method for the electro-thermal interconnected virtual power plant in the embodiment of the present invention;

[0084] Figure 2 It is a system composition diagram of the electro-thermal interconnected virtual power plant in Embodiment 2 of the present invention;

[0085] Figure 3 It is a comparison diagram of the convergence curves of different algorithms in Embodiment 2 of the present invention. Specific Embodiments

[0086] The present invention will be further described below in conjunction with specific embodiments.

[0087] Embodiment 1

[0088] An economic dispatch solution method for an electro-thermal interconnected virtual power plant, as Figure 1 shown, includes the following steps:

[0089] S1: Obtain the information of the equipment in the virtual power plant to be solved and the time-of-use electricity price on the current day, and establish the main optimization objective function according to the type of unit to be solved;

[0090] S2: Decouple the main optimization objective function in step S1 according to the peak-valley electricity price to obtain the first sub-objective function;

[0091] S3: Decouple the electrical and thermal components of the electrical-thermal interconnection equipment in the first sub-objective function in step S2 to obtain the second sub-objective functions reflecting the power grid and the thermal grid of the virtual power plant to be solved respectively;

[0092] S4: Based on the Java intelligent agent development framework, establish a distributed multi-agent system. Use the improved cross-over algorithm to solve the second sub-objective function in step S3 to obtain the sub-population and distributed sub-optimizers. Arrange the distributed sub-optimizers in the distributed multi-agent system, replace each distributed sub-optimizer with an intelligent agent, and classify different intelligent agents;

[0093] S5: The intelligent agents respectively execute the improved horizontal crossover and improved vertical crossover, and use the improved cross-over optimization algorithm for solution;

[0094] S6: The intelligent agents use the same-kind crossover and swap strategy for update;

[0095] S7: Each distributed sub-optimizer uses the global optimal particle of the updated distributed sub-optimizer in step S6 as the initialization data for population update and stores the global optimal population and the corresponding global optimal fitness value;

[0096] S8: When the optimization iteration times are met, output the optimal population information of the virtual power plant economic dispatch solution optimization.

[0097] For the above economic dispatch solution method of the electrical-thermal interconnected virtual power plant, the main optimization objective function is determined according to the type of units to be solved. Based on the peak-valley electricity price, the time scale of the electrical-thermal interconnected virtual power plant is divided. At the same time, the electrical-thermal decoupling method is used to decentralize and autonomize the power transmission network and the heat transmission network of the virtual power plant, so as to achieve the double decoupling of the electrical-thermal interconnected virtual power plant. Then, the second sub-objective function is calculated based on the divided sub-problems. Through two-dimensional reduction, the local optimal defect of the large-scale variable optimization problem is avoided, and the convergence accuracy is improved. The sub-objective multi-population interaction operation co-evolves, and the improved cross-over optimization algorithm and the same-kind intelligent agent crossover and swap strategy are used to solve the sub-problems, accelerating the convergence speed and improving the ability to mine detailed information of the convergence algorithm. Finally, when the iteration times are met, output the optimal population information of the virtual power plant economic dispatch solution optimization. In this embodiment, the heat and power grids only need to exchange a small amount of boundary variables and the virtual target factors are globally adjusted respectively to obtain the multi-objective solution, reducing the communication burden and the difficulty of problem solving.

[0098] In step S1, the main optimization objective function is:

[0099]

[0100] In the formula, represents the electricity sales volume of the virtual power plant to the system at the t-th moment; represents the electricity selling price at the \(t\)-th moment; represents the electricity purchase quantity of the virtual power plant from the system at the \(t\)-th moment; represents the electricity purchase price at the \(t\)-th moment; pen t represents the penalty coefficient at the \(t\)-th moment; represents the fuel cost at the \(t\)-th moment; represents the operation and maintenance cost at the \(t\)-th moment; N t represents the total time period, \(t\in[1,2,\cdots,N t .

[0101] In step S2, according to the peak, valley, and normal time periods corresponding to the time-of-use electricity price, the main optimization objective function in step S1 is decoupled and divided into N a peak moments, N v valley moments, N f normal moments, where N a , N v and N f are all less than N t and N a +N v +N f =N t ;

[0102] The first sub-objective function at peak time after decoupling is:

[0103]

[0104] In the formula, represents the first sub-objective function at peak time after time decoupling; represents the electricity selling quantity of the virtual power plant to the system at the \(t\)-th a moment; represents the electricity selling price at the \(t\)-th a moment; represents the electricity purchase quantity of the virtual power plant from the system at the \(t\)-th a moment; represents the electricity purchase price at the \(t\)-th a moment; represents the penalty coefficient at the \(t\)-th a moment; represents the fuel cost at the \(t\)-th a moment; represents the operation and maintenance cost at the \(t\)-th a moment; \(\alpha\) p represents the electricity coefficient of the combined heat and power unit; \(\alpha\) h represents the heat coefficient of the combined heat and power unit; represents the heat generation quantity of the combined heat and power unit at the \(t\)-th a moment; eff bl represents the energy efficiency coefficient of the boiler; Indicates the calorific value of the boiler at the t-th a moment; Q gas Indicates the price of natural gas; Indicates the t-th a moment's operation and maintenance cost of the energy storage device; Indicates the t-th a moment's power generation of the combined heat and power unit; Z pchp Indicates the operation and maintenance cost per unit of power generated by the combined heat and power unit; Z hchp Indicates the operation and maintenance cost per unit of heat generated by the combined heat and power unit; Indicates the power generation of the photovoltaic at the ta-th moment; Z ppv Indicates the operation and maintenance cost per unit of power generated by the photovoltaic; Indicates the t-th a moment's power generation of the wind turbine; Z pw Indicates the operation and maintenance cost per unit of power generated by the wind turbine; Z bl Indicates the operation and maintenance cost per unit of heat generated by the boiler;

[0105] The first sub-objective function during the valley period after decoupling is:

[0106]

[0107]

[0108] In the formula, Indicates the first sub-objective function during the valley period after time decoupling; Indicates the t-th v moment's power sold by the virtual power plant to the system; Indicates the t-th v moment's power selling price; Indicates the t-th v moment's power purchased by the virtual power plant from the system; Indicates the t-th v moment's power purchase price; Indicates the t-th v moment's penalty coefficient; Indicates the t-th v moment's fuel cost; Indicates the t-th v moment's operation and maintenance cost; Indicates the t-th v moment's calorific value of the combined heat and power unit; Indicates the t-th v moment's calorific value of the boiler; Indicates the t-th v moment's operation and maintenance cost of the energy storage device; Indicates the t-th v moment's power generation of the combined heat and power unit; represents the power generation of the photovoltaic at the t-th v moment; represents the power generation of the wind turbine at the t-th v moment;

[0109] The first sub-objective function in normal times after decoupling is:

[0110]

[0111] In the formula, represents the first sub-objective function in normal times after time decoupling; represents the electricity sales volume of the virtual power plant to the system at the t-th f moment; represents the electricity sales price at the t-th f moment; represents the electricity purchase volume of the virtual power plant from the system at the t-th f moment; represents the electricity purchase price at the t-th f moment; represents the penalty coefficient at the t-th f moment; represents the fuel cost at the t-th f moment; represents the operation and maintenance cost at the t-th f moment; represents the heat output of the combined heat and power unit at the t-th f moment; represents the heat output of the boiler at the t-th f moment; represents the operation and maintenance cost of the energy storage device at the t-th f moment; represents the power generation of the combined heat and power unit at the t-th f moment; represents the power generation of the photovoltaic at the t-th f moment; represents the power generation of the wind turbine at the t-th f moment.

[0112] In step S3, the electro-thermal decoupling of the electro-thermal interconnection device in the first sub-objective function in step S2 is performed to obtain:

[0113] The second sub-objective function of the power grid during peak hours is:

[0114]

[0115] The second sub-objective function of the heat network during peak hours is:

[0117]

[0118] The second sub-objective function of the off-peak power grid is:

[0119]

[0120] The second sub-objective function of the off-peak heat network is:

[0121]

[0122] The second sub-objective function of the normal-time power grid is:

[0123]

[0124] The second sub-objective function of the normal-time heat network is:

[0125]

[0126] In step S4, the sub-population corresponding to the second sub-objective function of the peak-time power grid The expression is:

[0127] The sub-population corresponding to the second sub-objective function of the peak-time heat network The expression is:

[0128]

[0129] The sub-population corresponding to the second sub-objective function of the off-peak power grid The expression is:

[0130]

[0131] The sub-population corresponding to the second sub-objective function of the off-peak heat network The expression is:

[0132]

[0133] The sub-population corresponding to the second sub-objective function of the normal-time power grid The expression is:

[0134]

[0135] The sub-population corresponding to the second sub-objective function of the normal-time heat network The expression is:

[0136]

[0137] The population expression corresponding to the global total population X is as follows:

[0138]

[0139] The agents are divided into two categories, where:

[0140] The expression of one type of agent is:

[0141]

[0142] In the formula; represents the sub-optimizer for solving the sub-objective function of the power grid during peak hours; represents using the CSO optimization algorithm to solve the sub-objective function of the power grid during peak hours; represents the sub-optimizer for solving the sub-objective function of the power grid during valley hours; represents using the CSO optimization algorithm to solve the sub-objective function of the power grid during valley hours; represents the sub-optimizer for solving the sub-objective function of the power grid during normal hours; represents using the CSO optimization algorithm to solve the sub-objective function of the power grid during normal hours;

[0143] The expression of the second type of agent is:

[0144]

[0145] In the formula; represents the sub-optimizer for solving the sub-objective function of the heat network during peak hours; represents using the CSO optimization algorithm to solve the sub-objective function of the heat network during peak hours; represents the sub-optimizer for solving the sub-objective function of the heat network during valley hours; represents using the CSO optimization algorithm to solve the sub-objective function of the heat network during valley hours; represents the sub-optimizer for solving the sub-objective function of the heat network during normal hours; represents using the CSO optimization algorithm to solve the sub-objective function of the heat network during normal hours.

[0146] In step S5, a random differential mutation strategy is adopted inside each distributed sub-optimizer for horizontal crossover:

[0147]

[0148]

[0149] In the formula, respectively represent the solutions after horizontal crossover of the agents corresponding to each second sub-objective function; j and k respectively represent the particles of the agent population;

[0150]

[0151] respectively represent the solutions randomly generated by the agents corresponding to each second sub-objective function; respectively represent all dimensions of the j-th particle of the corresponding sub-optimizer; respectively represent all dimensions of the k-th particle of the corresponding sub-optimizer; r and n represent random numbers uniformly distributed between [0, 1]; c represents a random number uniformly distributed between [-1, 1].

[0152] In step S5, vertical crossover is performed inside each distributed sub-optimizer:

[0153]

[0154] In the formula, respectively represent the solutions after vertical crossover of the agents corresponding to each second sub-objective function; d1 and d2 respectively represent the d1-th dimension and the d2-th dimension, d1≠d2 and d2∈(1, D), where D represents the total dimension of the problem to be solved.

[0155] In step S6, the expression of the same-kind crossover and swap strategy is:

[0156]

[0157] In the formula, respectively represent the solutions after the same-kind crossover and swap of the corresponding sub-optimizer; respectively represent all dimensions of the z-th particle of the corresponding sub-optimizer; respectively represent all dimensions of the ν-th particle of the corresponding sub-optimizer; respectively represent all dimensions of the u-th particle of the corresponding sub-optimizer; r1 and r2 respectively represent random numbers uniformly distributed between [0, 1].

[0158] Population update includes:

[0159] S701: Each distributed sub-optimizer calculates the second sub-objective function respectively and updates the population according to the greedy extremum to obtain sub-global optimal particles respectively

[0160] S702: Obtain the total global optimal particle

[0161] S703: Calculate the global optimal fitness.

[0162] Embodiment 2

[0163] In this embodiment, a certain virtual power plant is taken as an example. As Figure 2 shown, it includes a combined heat and power unit with a capacity of 600 kW, a boiler with a capacity of 500 kW, a wind power plant containing a fan with a capacity of 300 kW, a photovoltaic power station containing a photovoltaic system with a capacity of 305 kW, and four electric vehicles. The traditional CSO (Chicken Swarm Optimization) algorithm, the distributed CSO algorithm, and the economic dispatch solution method of the electro-thermal interconnected virtual power plant in Embodiment 1 are respectively used for experiments. The convergence curves of different algorithms are as Figure 3 shown. The convergence curve of the traditional CSO algorithm shows a slow downward trend. Compared with the traditional CSO algorithm, the convergence curve in Embodiment 1 shows a steeper trend, with a faster descent speed than the traditional CSO algorithm and the distributed CSO algorithm, and a higher convergence accuracy.

[0164] In the specific content of the above specific implementation manner, each technical feature can be combined arbitrarily without contradiction. For the sake of brevity of description, not all possible combinations of the above technical features are described. However, as long as the combinations of these technical features do not conflict, they should all be considered to be within the scope described in this specification.

[0165] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, rather than limitations on the implementation manners of the present invention. For those of ordinary skill in the art, other different forms of changes or modifications can be made on the basis of the above description. It is not necessary and impossible to list all the implementation manners here. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the claims of the present invention.

Claims

1. A method for solving the economic dispatch of an electric heating interconnected virtual power plant, characterized in that It includes the following steps: S1: Obtain the information of the equipment in the virtual power plant to be solved and the time-of-use electricity price of the current day, and establish the main optimization objective function according to the type of unit to be solved; S2: Decouple the main optimization objective function in step S1 according to the peak-valley electricity price to obtain the first sub-objective function; S3: Decouple the electro-thermal interconnected equipment in the first sub-objective function in step S2 to obtain the second sub-objective functions reflecting the power grid and the heat network of the virtual power plant to be solved respectively; S4: Based on the Java intelligent agent development framework, establish a distributed multi-agent system, use the improved cross and vertical algorithm to solve the second sub-objective function in step S3 to obtain the sub-population and the distributed sub-optimizer, arrange the distributed sub-optimizer in the distributed multi-agent system, replace each distributed sub-optimizer with an intelligent agent, and classify different intelligent agents; S5: The improved horizontal crossover and the improved vertical crossover are respectively executed inside the intelligent agent, and the improved cross and vertical optimization algorithm is used for solution; S6: The same-kind crossover and exchange strategy is adopted for updating among intelligent agents; S7: Each distributed sub-optimizer uses the global optimal particle of the updated distributed sub-optimizer in step S6 as the initialization data for population update and stores the global optimal population and the corresponding global optimal fitness value; S8: Meet the optimization iteration times and output the optimal population information of the economic dispatch solution of the virtual power plant.

2. The economic dispatch solution method for the electric-heat interconnected virtual power plant according to claim 1, wherein In step S1, the main optimization objective function is as follows: Wherein, represents the electricity sales volume of the virtual power plant to the system at the t-th moment; represents the electricity sales price at the t-th moment; represents the electricity purchase volume of the virtual power plant from the system at the t-th moment; represents the electricity purchase price at the t-th moment; pen t represents the penalty coefficient at the t-th moment; represents the fuel cost at the t-th moment; represents the operation and maintenance cost at the t-th moment; N t represents the total score moment, t ∈ [1, 2, ……, N t .

3. The economic dispatch solution method of the electric-heat interconnected virtual power plant according to claim 2, characterized in that, In step S2, the main optimization objective function in step S1 is decoupled and divided into N a peak moments, N v valley moments, and N f flat moments according to the peak-valley-flat periods corresponding to the time-of-use electricity price, where N a , N v , and N f are all less than N t and N a + N v + N f = N t ; The first sub-objective function at peak time obtained after decoupling is: In the formula, represents the first sub-objective function of the peak time after time decoupling; represents the electricity sales volume of the virtual power plant to the system at the a t-th represents the electricity sales price at the a t-th represents the electricity purchase volume of the virtual power plant from the system at the a t-th represents the electricity purchase price at the a t-th represents the penalty coefficient at the a t-th represents the fuel cost at the a t-th represents the operation and maintenance cost at the a t-th; α p represents the electricity coefficient of the combined heat and power unit; α h represents the heat coefficient of the combined heat and power unit; represents the heat output of the combined heat and power unit at the a t-th; eff bl represents the energy efficiency coefficient of the boiler; represents the heat output of the boiler at the a t-th; Q gas represents the price of natural gas; represents the operation and maintenance cost of the energy storage device at the a t-th; represents the electricity generation of the combined heat and power unit at the a t-th; Z pchp represents the operation and maintenance cost per unit of electricity generated by the combined heat and power unit; Z pchp represents the operation and maintenance cost per unit of heat generated by the combined heat and power unit; represents the electricity generation of the photovoltaic at the a t-th; Z ppv represents the operation and maintenance cost per unit of electricity generated by the photovoltaic; represents the electricity generation of the wind turbine at the a t-th; Z pw represents the operation and maintenance cost per unit of electricity generated by the wind turbine; Z bl represents the operation and maintenance cost per unit of heat generated by the boiler; The first sub-objective function at valley time obtained after decoupling is: In the formula, represents the first sub-objective function during the valley period after time decoupling; represents the electricity sales volume of the virtual power plant to the system at the v t-th moment; represents the electricity sales price at the v t-th moment; represents the electricity purchase volume of the virtual power plant from the system at the v t-th moment; represents the electricity purchase price at the v t-th moment; represents the penalty coefficient at the v t-th moment; represents the fuel cost at the v t-th moment; represents the operation and maintenance cost at the v t-th moment; represents the heat output of the combined heat and power unit at the v t-th moment; represents the heat output of the boiler at the v t-th moment; represents the operation and maintenance cost of the energy storage device at the v t-th moment; represents the electricity generation of the combined heat and power unit at the v t-th moment; represents the electricity generation of the photovoltaic at the v t-th moment; represents the electricity generation of the wind turbine at the v t-th moment; The first sub-objective function at normal time obtained after decoupling is: In the formula, represents the first sub-objective function during normal times after time decoupling; represents the electricity sales volume of the virtual power plant to the system at the f t-th moment; represents the electricity sales price at the f t-th moment; represents the electricity purchase volume of the virtual power plant from the system at the f t-th moment; represents the electricity purchase price at the f t-th moment; represents the penalty coefficient at the f t-th moment; represents the fuel cost at the f t-th moment; represents the operation and maintenance cost at the f t-th moment; represents the heat output of the combined heat and power unit at the f t-th moment; represents the heat output of the boiler at the f t-th moment; represents the operation and maintenance cost of the energy storage device at the f t-th moment; represents the electricity generation of the combined heat and power unit at the f t-th moment; represents the electricity generation of the photovoltaic at the f t-th moment; represents the electricity generation of the wind turbine at the f t-th moment.

4. The economic dispatch solution method of the electric-heat interconnected virtual power plant according to claim 2, wherein In step S3, the electro-thermal interconnected equipment in the first sub-objective function in step S2 is decoupled electro-thermally to obtain: The second sub-objective function of the peak-time power grid is as follows: The second sub-objective function of the peak-time heat supply network is as follows: The second sub-objective function of the valley-time power grid is as follows: The second sub-objective function of the off-peak heating network is as follows: The second sub-objective function of the power grid in normal times is as follows: The second sub-objective function of the normal heat supply network is as follows:

5. The economic dispatch solution method for the electric-heat interconnected virtual power plant according to claim 4, wherein, In step S4, the sub-population corresponding to the second sub-objective function of the peak-time power grid The expression is: The sub-population corresponding to the second sub-objective function of the peak-time heat supply network The expression is: The subpopulation corresponding to the second sub-objective function of the valley-time power grid The expression is: The sub-population corresponding to the second sub-objective function of the off-peak heating network The expression is: The sub-population corresponding to the second sub-objective function of the power grid in normal times The expression is: The sub-population corresponding to the second sub-objective function of the normal heating network The expression is: The population expression corresponding to the global total population X is:

6. The economic dispatch solution method of the electric-heat interconnected virtual power plant according to claim 5, characterized in that In step S4, the intelligent agents are divided into two categories, where: The expression of one type of intelligent agent is: where; represents the sub-optimizer for solving the sub-objective function of the power grid during peak hours; represents using the CSO optimization algorithm to solve the sub-objective function of the power grid during peak hours; represents the sub-optimizer for solving the sub-objective function of the power grid during valley hours; represents using the CSO optimization algorithm to solve the sub-objective function of the power grid during valley hours; represents the sub-optimizer for solving the sub-objective function of the power grid during normal hours; represents using the CSO optimization algorithm to solve the sub-objective function of the power grid during normal hours; The expression of the second type of intelligent agent is: In the formula; represents the sub-optimizer for the sub-objective function of the heat supply network when solving for the peak; represents using the CSO optimization algorithm to solve the sub-objective function of the heat supply network at the peak; represents the sub-optimizer for the sub-objective function of the heat supply network when solving for the valley; represents using the CSO optimization algorithm to solve the sub-objective function of the heat supply network at the valley; represents the sub-optimizer for the sub-objective function of the heat supply network when solving for normal times; represents using the CSO optimization algorithm to solve the sub-objective function of the heat supply network at normal times.

7. The economic dispatch solution method for the electric-heat interconnected virtual power plant according to claim 6, wherein In step S5, a strategy of random differential mutation is adopted inside each distributed sub-optimizer for horizontal crossover: respectively represent the solutions after horizontal crossover of the agents corresponding to each second sub-objective function; j and k respectively represent the particles of the agent population; respectively represent the solutions randomly generated by the agents corresponding to each second sub-objective function; respectively represent all dimensions of the j-th particle of the corresponding sub-optimizer; respectively represent all dimensions of the k-th particle of the corresponding sub-optimizer; r and n represent random numbers uniformly distributed between [0, 1]; c represents a random number uniformly distributed between [-1, 1].

8. The economic dispatch solution method for the electric-heat interconnected virtual power plant according to claim 7, characterized in that In step S5, vertical crossover is performed inside each distributed sub-optimizer: wherein, respectively represent the solutions after vertical crossover of the agents corresponding to each second sub-objective function; d1 and d2 respectively represent the d1-th dimension and the D2-th dimension, where D1≠D2 and d2∈(1, D), and D represents the total dimension of the problem to be solved.

9. The economic dispatch solution method for the electric-heat interconnected virtual power plant according to claim 8, characterized in that, In step S6, the expression of the same-kind crossover and exchange strategy is: In the formula, respectively represent the solutions after the corresponding sub-optimizers perform homogeneous crossover and swap; respectively represent all dimensions of the z-th particle of the corresponding sub-optimizer; respectively represent all dimensions of the ν-th particle of the corresponding sub-optimizer; respectively represent all dimensions of the u-th particle of the corresponding sub-optimizer; r1 and r2 respectively represent random numbers uniformly distributed between [0, 1].

10. The economic dispatch solution method for the electric-heat interconnected virtual power plant according to claim 8, wherein Population update includes: S701: Each distributed sub-optimizer calculates the second sub-objective function respectively, updates the population according to the greedy extreme value, and obtains the sub-global optimal particles respectively. S702: Obtain the overall global optimal particle S703: Calculate the global optimal fitness.

Images