Power distribution network-agricultural park double-layer optimization method and system

By building a two-layer optimization model and improved algorithm for distribution network-agricultural parks, the decision-making conflict between distribution network operators and agricultural park users was solved, the pricing of distribution network operators and the power consumption strategy of agricultural parks was optimized, and the power distribution was reduced. The goal of network operation costs and improving the user benefits of agricultural parks.

CN118982092APending Publication Date: 2024-11-19STATE GRID SICHUAN ECONOMIC RES INST +1
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410958611.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

The existing technology lacks decision-making optimization methods for distribution network operators and agricultural park users, resulting in low power supply reliability and large peak-to-valley difference, making it difficult to effectively coordinate the power consumption strategies of distributed photovoltaics and agricultural facilities, and adding the power grid. Running cost.

Method used

Build a distribution network-agricultural park double-layer optimization model, combine the improved particle swarm algorithm and cplex solver, obtain the pricing strategy of distribution network operators and agricultural park power consumption strategies, and optimize the interaction between distribution network operators and agricultural park users , reduce the operating costs of distribution networks and improve the user benefits of agricultural parks.

Benefits of technology

Through the two-layer optimization model and algorithm solution, the optimal distribution network operator pricing strategy and agricultural park user power consumption strategy are obtained, which meets the safe power supply and user needs of the power grid, reduces the operating costs of the distribution network and increases agricultural park users. benefit.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982092A_ABST
    Figure CN118982092A_ABST
Patent Text Reader

Abstract

The invention discloses a power distribution network-agricultural park double-layer optimization method and system, relates to the technical field of power distribution network planning, and solves the problem that a decision optimization method for a power distribution network operator and an agricultural park user does not exist at present. Basic data such as a power distribution network topological structure, distributed photovoltaic and park agricultural facility parameters are obtained; s2, constructing a mathematical model for distributed photovoltaic power generation, power utilization in the greenhouse planting industry and power utilization in the animal product breeding industry; s3, considering decision conflicts of a power distribution network operator and agricultural park users in benefit improvement, and constructing a power distribution network-agricultural park double-layer optimization model; and S4, solving the model established in the S3 by adopting an improved particle swarm algorithm in combination with a cplex solver. According to the method, the interaction between the power distribution network operator and the agricultural park with adjustable resources is fully considered, and the benefits of the users in the agricultural park are effectively improved while the operation cost of the power distribution network is reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network planning, and more specifically, to a distribution network-agricultural park double-layer optimization method and system. Background Art

[0003] The infrastructure and technology of power grids in rural areas of my country are facing risks such as low power supply reliability, insufficient power supply capacity of distribution networks and large peak-to-valley differences. With the high penetration of distributed photovoltaics into rural distribution networks, the intermittent and volatile characteristics of distributed photovoltaics have brought greater challenges to the peak regulation, frequency regulation, power quality, power supply reliability, and photovoltaic consumption of rural distribution networks. In recent years, agricultural models dominated by greenhouse planting and livestock breeding have been widely popularized in rural areas. However, greenhouses and livestock and poultry houses need to consume a lot of energy to control and improve environmental factors to achieve high yields. These two high-energy loads have a great impact on the safe and reliable operation of distribution networks in rural areas. At present, there is no decision-making optimization method for distribution network operators and agricultural park users to guide the pricing strategy of distribution network operators to match the electricity consumption strategy of agricultural park users, so as to improve the benefits of agricultural park users and reduce the operating costs of distribution networks.

[0004] Therefore, the present application provides a distribution network-agricultural park double-layer optimization method and system to solve the above problems. Summary of the invention

[0005] The purpose of this application is to provide a distribution network-agricultural park two-layer optimization method and system to solve the problem that there is currently no decision-making optimization method for distribution network operators and agricultural park users, and to guide the distribution network operator's pricing strategy to coordinate with the agricultural park user's electricity consumption strategy, so as to achieve the improvement of agricultural park user benefits while reducing the distribution network operation cost; this application constructs a distribution network-agricultural park two-layer optimization model, analyzes the interaction between distribution network operators and agricultural parks, and provides mutually coordinated distribution network operator pricing strategies and agricultural park electricity consumption strategies, so as to reduce the distribution network operation cost while improving the benefits of agricultural park users.

[0006] The present application first provides a distribution network-agricultural park double-layer optimization method, comprising: step S1, obtaining basic data, including: distribution network topology, distributed photovoltaic and park agricultural facility parameters; step S2, based on the basic data, constructing a mathematical model of distributed photovoltaic power generation, greenhouse planting electricity consumption and livestock breeding electricity consumption; step S3, based on the data model, considering the decision-making conflict between distribution network operators and agricultural park users on benefit improvement, and constructing a distribution network-agricultural park double-layer optimization model; step S4, using an improved particle swarm algorithm combined with a cplex solver to solve the distribution network-agricultural park double-layer optimization model to obtain the optimal distribution network operator pricing strategy and the optimal agricultural park user electricity consumption strategy.

[0007] By adopting the above technical scheme and considering the coordination between distributed photovoltaic and agricultural facilities, a distribution network-agricultural park two-layer optimization model is constructed to reflect the decision-making conflict between distribution network operators and agricultural park users in improving benefits. The model is solved by combining the improved particle swarm algorithm with the cplex solver to obtain the optimal pricing strategy of distribution network operators and the optimal electricity consumption strategy of agricultural park users. The interaction between distribution network operators and agricultural parks with adjustable resources is fully considered to solve the corresponding pricing strategy of distribution network operators and electricity consumption strategy of agricultural park users to meet the requirements of safe power supply of distribution network and the electricity demand of agricultural park users.

[0008] In a possible implementation, step S1 includes: obtaining the distribution network topology, transformer equipment parameters, switch position, line parameters, distributed photovoltaic access location, capacity, power parameters, the number and equipment configuration of greenhouses in the agricultural park, and the number and equipment configuration of livestock farms.

[0009] In a possible implementation, step S2 includes:

[0010] The specific mathematical model of distributed photovoltaic power generation is:

[0011]

[0012]

[0013]

[0014]

[0015] Where: P is the set of distributed photovoltaic access nodes; is the connection relationship between distributed photovoltaic and distribution network node i; and They are distributed photovoltaic active power and reactive power respectively; and They are respectively the maximum value of distributed photovoltaic active power, the maximum value of reactive power and the minimum value of reactive power; is the capacity of distributed photovoltaic inverter at node i;

[0016] The specific mathematical model of electricity consumption in greenhouse planting industry and livestock breeding industry is:

[0017]

[0018]

[0019]

[0020]

[0021]

[0022] Where: is the electricity consumption of greenhouse planting industry at time t; and are the power consumption of irrigation, ventilation and lighting equipment in the greenhouse at time t; and is a Boolean variable. If it is 1, it means that the irrigation, ventilation and supplementary lighting equipment in the greenhouse are put into use at time t. n1, n2 and n3 respectively represent the number of irrigation, ventilation and supplementary lighting equipment in the greenhouse; is the power consumption of livestock breeding industry at time t; k is the total number of types of equipment in livestock breeding farms; is a Boolean variable. If it is 1, it means that the farm equipment n is put into use at time t; n is the number of farm equipment n; P n is the rated power of the farm equipment n; m water The actual amount of irrigation water required in the greenhouse every day; is the amount of irrigation water required for the greenhouse every day under ideal conditions; j and They are the influencing factors and change rates of irrigation water volume; is the efficiency parameter of the irrigation equipment; g is the acceleration of gravity; G water is the amount of water pumped; It refers to the usage time of irrigation equipment within 24 hours a day.

[0023] In a possible implementation, step S3 includes: taking the pricing strategy of the distribution network operator as a decision variable, considering the peak-to-valley difference rate, power supply cost, abandoned light cost and network loss cost to construct the comprehensive operating cost of the distribution network operator, taking reducing the comprehensive operating cost of the distribution network operator as the optimization goal, and constructing an upper-level model; based on the pricing of the distribution network operator output by the upper-level model, taking the electricity consumption of the greenhouse planting industry and the electricity consumption strategy of the livestock breeding industry as decision variables, and taking maximizing the benefits of agricultural park users as the optimization goal, constructing a lower-level model.

[0024] In a possible implementation, the upper model is:

[0025] F=C FG +C BG +C PV +C SL +C DR

[0026]

[0027]

[0028]

[0029]

[0030] Where: F is the comprehensive operating cost of the distribution network operator; C FG , C BG , C PV , C SL , C DR are the peak-valley difference smoothing cost, power supply guarantee cost, abandoned light cost, network loss cost and demand response compensation cost of the distribution network operator; c FG N is the cost coefficient for smoothing the peak-to-valley difference of the distribution network; N is the total number of nodes; P t,i is the net load value of node i in period t, the load is defined as positive and the power output is defined as negative; c BG Guaranteed supply cost per unit of electricity; is the loss load required to be guaranteed for node j in time period t; c PV The cost of curtailing solar power per unit of electricity; is the predicted maximum output value of distributed photovoltaic j in time period t; is the output value of distributed photovoltaic j called by the distribution network operator in time period t; c SL is the network loss cost coefficient; r ij is the resistance value between lines ij; I ij,t is the square value of the current of line ij in time period t; L is the line set.

[0031] In a possible implementation, the lower layer model is:

[0032]

[0033] Where: is the basic electricity price charged by the distribution network operator at time t; N f is the set of nodes where the agricultural park is located; is the predicted basic power consumption of greenhouse planting industry in agricultural park at time t; is the predicted basic power consumption of livestock breeding industry at time t; The demand response electricity price compensation strategy formulated by the distribution network operator at time t; The actual power consumption of greenhouse planting industry in the agricultural park at time t after real-time demand response strategy; The actual electricity consumption of the livestock breeding industry at time t after the real-time demand response strategy.

[0034] In a possible implementation, the constraints of the distribution network-agricultural park two-layer optimization model are:

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042]

[0043] Where: P ij is the active power flowing between nodes ij; Q ij is the reactive power flowing between nodes ij; R ij is the resistance value between the distribution network lines ij; X ij is the reactance value between the distribution network lines ij; P L i,max is the maximum value of active power of load node i; Q L i,max is the maximum value of reactive power of load node i; is the square of the maximum voltage of load node i in the distribution network; is the square of the minimum voltage of load node i in the distribution network; is the square of the maximum current between the distribution network lines ij; ω ij is the Boolean variable of the connection of the ij node; ω ij =1 means that node j is connected; M is a very large positive number.

[0044] In a possible implementation, the improved particle swarm algorithm in step S4 is as follows: step A, initializing the population; step B, calculating the fitness value of each particle in the population; step C, updating the inertia weight according to an adaptive linear descent method, and sorting the population particles according to the fitness value; step D, adopting a role model learning strategy for non-optimal particles in the population, adopting a random reverse learning strategy for the optimal particles in the population in the early stage of iteration, and adopting a learning strategy combining general reverse learning and adaptive mutation mechanism in the late stage of iteration; step E, calculating new fitness values ​​in parallel, selecting the optimal particles and the global optimum; step F, judging whether the termination condition is met, if so, terminating the iteration, if not, returning to step C.

[0045] In a possible implementation, step S4 includes: step S41, inputting the decision variables of the upper model in the distribution network-agricultural park two-layer optimization model, setting the particle swarm algorithm parameters, and setting the number of iterations k=0; step S42, using the particle swarm algorithm to randomly generate m groups of distribution network operator pricing, and passing the parameters to the lower model; step S43, the lower model targets the distribution network operator pricing, calls the cplex solver to calculate the optimal greenhouse planting industry electricity consumption and livestock breeding industry electricity consumption strategy with the goal of maximizing benefits, retains the current agricultural park user benefits and returns the electricity consumption strategy to the upper model; step S44, the upper model is based on the user's greenhouse planting The electricity consumption strategies of the agricultural industry and the livestock breeding industry are calculated and retained, and the current distribution network operator's revenue and pricing strategy are calculated and retained; step S45, recording the particle position of the m groups of pricing, the optimal position of the population, the particle fitness and the optimal fitness of the population; step S46, if the number of iterations is less than the set number of iterations, the particle swarm algorithm is used to generate new m groups of distribution network operator pricing and jump to step S43, and the number of iterations k=k+1; if the number of iterations is greater than or equal to the set number of iterations, step S47 is executed; step S47, output the optimal distribution network operator pricing strategy, the optimal agricultural park user electricity consumption strategy, the distribution network operator's revenue and the agricultural park user's benefit.

[0046] The present application also provides a distribution network-agricultural park two-layer optimization system, which is used to execute a distribution network-agricultural park two-layer optimization method as described above, including: a data acquisition unit, used to acquire basic data, including: distribution network topology, distributed photovoltaic and park agricultural facility parameters; a basic model construction unit, used to construct a mathematical model of distributed photovoltaic power generation, greenhouse planting electricity consumption and livestock breeding electricity consumption based on the basic data; an optimization model construction unit, used to construct a distribution network-agricultural park two-layer optimization model based on the data model, considering the decision-making conflict between distribution network operators and agricultural park users on efficiency improvement; an optimization model solving unit, used to solve the distribution network-agricultural park two-layer optimization model using an improved particle swarm algorithm combined with a cplex solver, and obtain the optimal distribution network operator pricing strategy and the optimal agricultural park user electricity consumption strategy.

[0047] Compared with the prior art, the present application has the following beneficial effects: the present solution fully considers the interaction between the distribution network operator and the agricultural park with adjustable resources, solves the corresponding pricing strategy of the distribution network operator and the electricity consumption strategy of the agricultural park users, and meets the requirements of safe power supply of the distribution network and the electricity demand of the agricultural park users; constructs a two-layer optimization model with the optimization goals of reducing the comprehensive operating costs of the distribution network operator and maximizing the benefits of the agricultural park users, solves the pricing strategy of the distribution network operator and the electricity consumption strategy of the agricultural park users, which can guide the pricing of the distribution network operator, guide the electricity consumption of the agricultural park users, reduce the operating costs of the distribution network, and improve the benefits of the agricultural park users. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0049] Figure 1 It is a flow chart of a distribution network-agricultural park double-layer optimization method provided by an embodiment of the present invention;

[0050] Figure 2 It is a schematic diagram of an improved process of a particle swarm algorithm provided by an embodiment of the present invention;

[0051] Figure 3 It is a schematic diagram of a process for solving a distribution network-agricultural park double-layer optimization model based on an improved particle swarm algorithm combined with a cplex solver provided by the present invention;

[0052] Figure 4 A schematic diagram of the structure of a distribution network-agricultural park two-layer optimization system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with examples and drawings. The illustrative implementation scheme of the present application and its description are only used to explain the present application and are not intended to limit the present application.

[0054] See also Figure 1 As shown, Figure 1 A schematic flow chart of the distribution network-agricultural park dual-layer optimization method provided in the embodiment of the present application. The method includes the following:

[0055] Step S1, obtaining basic data, including: distribution network topology, distributed photovoltaic and park agricultural facility parameters; specifically, it may include: obtaining distribution network topology, transformer equipment parameters, switch position, line parameters, distributed photovoltaic access location, capacity, power parameters, the number of greenhouses in the agricultural park and equipment configuration, and the number of livestock farms and equipment configuration and other model calculation data.

[0056] Step S2: Based on the basic data, a mathematical model of distributed photovoltaic power generation, greenhouse planting electricity consumption and livestock breeding electricity consumption is constructed; specifically, it may be:

[0057] Mathematical model of distributed photovoltaic power generation:

[0058]

[0059]

[0060]

[0061]

[0062] Where: P is the set of distributed photovoltaic access nodes; is the connection relationship between distributed photovoltaic and distribution network node i; and They are distributed photovoltaic active power and reactive power respectively; and They are respectively the maximum value of distributed photovoltaic active power, the maximum value of reactive power and the minimum value of reactive power; is the capacity of distributed PV inverter at node i.

[0063] Mathematical model of electricity consumption in greenhouse planting industry and livestock breeding industry:

[0064]

[0065]

[0066]

[0067]

[0068]

[0069] Where: is the electricity consumption of greenhouse planting industry at time t; and are the power consumption of irrigation, ventilation and lighting equipment in the greenhouse at time t; and is a Boolean variable. If it is 1, it means that the irrigation, ventilation and supplementary lighting equipment in the greenhouse are put into use at time t. n1, n2 and n3 respectively represent the number of irrigation, ventilation and supplementary lighting equipment in the greenhouse; is the power consumption of livestock breeding industry at time t; k is the total number of types of equipment in livestock breeding farms; is a Boolean variable. If it is 1, it means that the farm equipment n is put into use at time t; n is the number of farm equipment n; P n is the rated power of the farm equipment n; m water The actual amount of irrigation water required in the greenhouse every day; is the amount of irrigation water required for the greenhouse every day under ideal conditions; j and They are the influencing factors and change rates of irrigation water volume; is the efficiency parameter of the irrigation equipment; g is the acceleration of gravity; G water is the amount of water pumped; It refers to the usage time of irrigation equipment within 24 hours a day.

[0070] Step S3: Based on the data model, considering the decision conflicts between the distribution network operator and the agricultural park users in terms of benefit improvement, a distribution network-agricultural park two-layer optimization model is constructed; specifically, it may be:

[0071] Taking the pricing strategy of distribution network operators as the decision variable, the comprehensive operating cost of distribution network operators is constructed by considering the peak-to-valley difference rate, power supply guarantee cost, abandoned light cost and network loss cost. The upper-level model is constructed with reducing the comprehensive operating cost of distribution network operators as the optimization goal. Based on the pricing of distribution network operators output by the upper-level model, the electricity consumption strategies of greenhouse planting industry and livestock breeding industry are taken as decision variables, and the maximum benefit of agricultural park users is taken as the optimization goal to construct the lower-level model.

[0072] It is understandable that the upper layer (distribution network operator layer) of the distribution network-agricultural park two-layer optimization model takes reducing the peak-to-valley difference rate, power supply cost, abandoned light cost and network loss cost as optimization goals, and minimizes the operating costs by calling on distributed photovoltaic power generation and other technical means and providing reasonable compensation prices to the lower layer (agricultural park user layer). In the lower layer of the model, agricultural park users change the electricity consumption strategies of greenhouse planting and livestock breeding industries based on the compensation prices set by distribution network operators and taking into account their actual production and living conditions with the objective function of maximizing self-generated benefits.

[0073] The upper layer (distribution network operator layer) of the two-layer optimization model of distribution network-agricultural park can be specifically modeled as follows:

[0074] F=C FG +C BG +C PV +C SL +C DR

[0075]

[0076]

[0077]

[0078]

[0079] Where: F is the comprehensive operating cost of the distribution network operator; C FG , C BG , C PV , C SL , C DR are the peak-valley difference smoothing cost, power supply guarantee cost, abandoned light cost, network loss cost and demand response compensation cost of the distribution network operator; c FG N is the cost coefficient for smoothing the peak-to-valley difference of the distribution network; N is the total number of nodes; P t,i is the net load value of node i in period t, the load is defined as positive and the power output is defined as negative; c BG Guaranteed supply cost per unit of electricity; is the loss load required to be guaranteed for node j in time period t; c PV The cost of curtailing solar power per unit of electricity; is the predicted maximum output value of distributed photovoltaic j in time period t; is the output value of distributed photovoltaic j called by the distribution network operator in time period t; c SL is the network loss cost coefficient; r ij is the resistance value between lines ij; I ij,t is the square value of the current of line ij in time period t; L is the line set.

[0080] The objective function of the lower layer (agricultural park user layer) of the two-layer optimization model of distribution network-agricultural park can be:

[0081]

[0082] Where: is the basic electricity price charged by the distribution network operator at time t; N f is the set of nodes where the agricultural park is located; is the predicted basic power consumption of greenhouse planting industry in agricultural park at time t; is the predicted basic power consumption of livestock breeding industry at time t; The demand response electricity price compensation strategy formulated by the distribution network operator at time t; The actual power consumption of greenhouse planting industry in the agricultural park at time t after real-time demand response strategy; The actual electricity consumption of the livestock breeding industry at time t after the real-time demand response strategy.

[0083] The constraints of the above distribution network-agricultural park two-layer optimization model can be:

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092] Where: P ij is the active power flowing between nodes ij; Q ij is the reactive power flowing between nodes ij; R ij is the resistance value between the distribution network lines ij; X ij is the reactance value between the distribution network lines ij; P L i,max is the maximum value of active power of load node i; Q L i,max is the maximum value of reactive power of load node i; is the square of the maximum voltage of load node i in the distribution network; is the square of the minimum voltage of load node i in the distribution network; is the square of the maximum current between the distribution network lines ij; ω ij is the Boolean variable of the connection of the ij node; ω ij =1 means that node j is connected; M is a very large positive number.

[0093] Step S4: using an improved particle swarm algorithm combined with a cplex solver to solve the distribution network-agricultural park two-layer optimization model to obtain the optimal distribution network operator pricing strategy and the optimal agricultural park user electricity consumption strategy.

[0094] It should be noted that the traditional particle swarm algorithm selects the optimal position p of a single particle best and the optimal position of the particle population g best 2 optimal positions to iteratively update, where each particle contains two influencing factors: velocity vector and position vector. Let the velocity vector of a particle m in D-dimensional space be Let the position vector of particle m be The iteration formula is as follows:

[0095]

[0096]

[0097] Where: c1 and c2 are the learning factors of the particle swarm algorithm; r1 and r2 are random numbers between zero and one; X m is the position of the mth particle in the particle swarm, and d represents the dimension number;

[0098] It can be seen that all particles in the population are close to the optimal position, but when the optimal position coincides with the local optimal point, the particle swarm algorithm tends to the local optimal solution. If the particle swarm algorithm is solving complex problems, because it contains multiple local optimal values, after multiple iterations, the population is extremely easy to fall into the vicinity of the local optimal solution, causing the global search performance of the population to deteriorate and the population diversity to decrease, and ultimately causing the particle swarm algorithm to converge to the local optimal solution.

[0099] Therefore, in order to improve the global optimization ability and population diversity of the particle swarm algorithm, this scheme combines example learning with reverse learning. The idea of ​​example learning is as follows:

[0100] The particles in the particle group are directed toward g best With p bestThe optimization of two positions is changed to a random optimization towards a better particle. Each non-optimal particle randomly selects a particle from a group of particles that are better than itself to learn, thereby improving its global search performance and effectively avoiding particles from being restricted to local optimal solutions. The pseudo code of this strategy is as follows:

[0101]

[0102]

[0103] Where: t is the number of iterations of the particle swarm algorithm, MaxD T is the maximum number of iterations of the particle swarm algorithm, N is the number of particle populations, m is the particle that is not the optimal solution (i = 2, 3, ..., N), c n is the sequence number of randomly selected particles, w t is the inertia weight of the particle swarm algorithm;

[0104] Since the best particles in the population have no role models for learning, this scheme selects random reverse learning in the early stage of iteration. The idea is as follows:

[0105]

[0106] in: is the global optimal particle corresponding to the maximum fitness value after each iteration; a and b are the boundaries of the particle; r is a random number between zero and one; the present invention selects a general reverse learning strategy in the later stage of iteration, and the idea is as follows:

[0107]

[0108] Finally, the present invention adds a mutation mechanism on the basis of general reverse learning, and selects an adaptive method for the mutation probability. The mutation probability will show a trend from large to small according to the increase in the number of iterations. Setting the mutation probability higher in the early stage of iteration can improve the algorithm's ability to search for the global optimal solution. Setting the mutation probability lower in the early and late stages of iteration can effectively improve the particle swarm algorithm's optimization speed. The algorithm pseudo code of reverse learning combined with the mutation mechanism is as follows:

[0109]

[0110]

[0111] Among them: the mutation probability value and the inertia weight w t Equal, inertia weight w t Select the adaptive linear descent method, the formula is as follows:

[0112] w t =1-t / MaxD T .

[0113] See also Figure 2 As shown in FIG. 1 , the improved particle swarm algorithm is specifically as follows: step A, initializing the population; step B, calculating the fitness value of each particle in the population; step C, updating the inertia weight according to the adaptive linear descent method, and sorting the population particles according to the fitness value; step D, adopting the example learning strategy for the non-optimal particles in the population, adopting the random reverse learning strategy for the optimal particles in the population in the early stage of iteration, and adopting the learning strategy combining the general reverse learning and the adaptive mutation mechanism in the late stage of iteration; step E, calculating the new fitness value in parallel, selecting the optimal particle and the global optimum; step F, judging whether the termination condition is met, if so, terminating the iteration, if not, returning to step C.

[0114] See also Figure 3 As shown, combined with the above-mentioned improved particle swarm algorithm, step S4 includes: step S41, input the decision variables of the upper model in the distribution network-agricultural park two-layer optimization model, set the particle swarm algorithm parameters, and set the number of iterations k=0; step S42, use the particle swarm algorithm to randomly generate m groups of distribution network operator pricing φ m , passing the parameters to the lower model; Step S43, the lower model sets the price φ for the distribution network operator m , calling the cplex solver to calculate the optimal electricity consumption strategy for greenhouse planting and livestock breeding with the goal of maximizing benefits, and retaining the current benefits of agricultural park users The power consumption strategy is returned to the upper model; Step S44, the upper model calculates and retains the current distribution network operator's revenue based on the user's greenhouse planting industry power consumption strategy and livestock breeding industry power consumption strategy and pricing strategy Step S45: Record the particle positions P of m groups of pricing md , the optimal position of the population P pd , particle fitness And the optimal fitness of the population F max ; Step S46, if the number of iterations is less than the set number of iterations k<k set , then use the particle swarm algorithm to generate new m groups of distribution network operator pricing and jump to step S43, set the number of iterations k = k + 1; if the number of iterations is greater than or equal to the set number of iterations k ≥ k set , then execute step S47; Step S47, output the optimal distribution network operator pricing strategy φ m , the optimal electricity strategy for agricultural park users F max , distribution network operator revenue max and benefits to users of agricultural parks.

[0115] See also Figure 4As shown, the present application also provides a distribution network-agricultural park two-layer optimization system for executing the distribution network-agricultural park two-layer optimization method as described above, including: a data acquisition unit for acquiring basic data, including: distribution network topology, distributed photovoltaic and park agricultural facility parameters; a basic model construction unit for constructing a mathematical model of distributed photovoltaic power generation, greenhouse planting electricity consumption and livestock breeding electricity consumption based on the basic data; an optimization model construction unit for constructing a distribution network-agricultural park two-layer optimization model based on the data model, considering the decision-making conflict between the distribution network operator and the agricultural park users on the benefit improvement; an optimization model solving unit for solving the distribution network-agricultural park two-layer optimization model using an improved particle swarm algorithm combined with a cplex solver to obtain the optimal distribution network operator pricing strategy and the optimal agricultural park user electricity consumption strategy.

[0116] It is understandable that this scheme considers the coordination of distributed photovoltaics and agricultural facilities, constructs a distribution network-agricultural park double-layer optimization model to reflect the decision-making conflict between distribution network operators and agricultural park users in improving benefits, and solves the model by combining the improved particle swarm algorithm with the cplex solver to obtain the optimal distribution network operator pricing strategy and the optimal agricultural park user electricity consumption strategy. Among them, the upper model of the distribution network-agricultural park double-layer optimization model takes reducing the comprehensive operating cost of the distribution network operator as the optimization goal, and the lower model takes maximizing the benefits of agricultural park users as the optimization goal. Through the model solution, the optimal distribution network operator pricing strategy and the optimal agricultural park user electricity consumption strategy that improve the benefits of agricultural park users while reducing the distribution network operation cost are obtained.

[0117] Compared with the existing technology, this solution fully considers the interaction between distribution network operators and agricultural parks with adjustable resources, and solves the corresponding pricing strategy of distribution network operators and electricity consumption strategy of agricultural park users to meet the requirements of safe power supply of distribution network and the electricity demand of agricultural park users; constructs a two-layer optimization model with the optimization goals of reducing the comprehensive operating costs of distribution network operators and maximizing the benefits of agricultural park users, and solves the pricing strategy of distribution network operators and electricity consumption strategy of agricultural park users, which can guide the pricing of distribution network operators, guide the electricity consumption of agricultural park users, reduce the operating costs of distribution networks, and improve the benefits of agricultural park users.

[0118] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A distribution network-agricultural park two-layer optimization method, characterized in that: include: Step S1, obtaining basic data, including: distribution network topology, distributed photovoltaic and park agricultural facility parameters; Step S2: constructing mathematical models of distributed photovoltaic power generation, electricity consumption in greenhouse planting and electricity consumption in livestock breeding based on the basic data; Step S3: Based on the data model, considering the decision conflicts between distribution network operators and agricultural park users on benefit improvement, a distribution network-agricultural park two-layer optimization model is constructed; Step S4: using an improved particle swarm algorithm combined with a cplex solver to solve the distribution network-agricultural park two-layer optimization model to obtain the optimal distribution network operator pricing strategy and the optimal agricultural park user electricity consumption strategy.

2. A distribution network-agricultural park double-layer optimization method according to claim 1, characterized in that: The step S1 includes: obtaining the distribution network topology, transformer equipment parameters, switch position, line parameters, distributed photovoltaic access location, capacity, power parameters, the number and equipment configuration of greenhouses in the agricultural park, and the number and equipment configuration of livestock farms.

3. A distribution network-agricultural park double-layer optimization method according to claim 1, characterized in that: The step S2 comprises: The specific mathematical model of distributed photovoltaic power generation is: Where: P is the set of distributed photovoltaic access nodes; is the connection relationship between distributed photovoltaic and distribution network node i; and They are distributed photovoltaic active power and reactive power respectively; and They are respectively the maximum value of distributed photovoltaic active power, the maximum value of reactive power and the minimum value of reactive power; is the capacity of distributed photovoltaic inverter at node i; The specific mathematical model of electricity consumption in greenhouse planting industry and livestock breeding industry is: Where: is the electricity consumption of greenhouse planting industry at time t; and are the power consumption of irrigation, ventilation and lighting equipment in the greenhouse at time t; and is a Boolean variable. If it is 1, it means that the irrigation, ventilation and supplementary lighting equipment in the greenhouse are put into use at time t. n1, n2 and n3 respectively represent the number of irrigation, ventilation and supplementary lighting equipment in the greenhouse; is the power consumption of livestock breeding industry at time t; k is the total number of types of equipment in livestock breeding farms; is a Boolean variable. If it is 1, it means that the farm equipment n is put into use at time t; n is the number of farm equipment n; P n is the rated power of the farm equipment n; m water The actual amount of irrigation water required in the greenhouse every day; is the amount of irrigation water required for the greenhouse every day under ideal conditions; j and They are the influencing factors and change rates of irrigation water volume; is the efficiency parameter of the irrigation equipment; g is the acceleration of gravity; G water is the amount of water pumped; It refers to the usage time of irrigation equipment within 24 hours a day.

4. A distribution network-agricultural park double-layer optimization method according to claim 1, characterized in that: The step S3 comprises: Taking the pricing strategy of distribution network operators as decision variables, the comprehensive operating cost of distribution network operators is constructed by considering the peak-to-valley difference rate, power supply guarantee cost, abandoned light cost and network loss cost. The upper-level model is constructed with the reduction of the comprehensive operating cost of distribution network operators as the optimization goal. The lower-level model is constructed based on the pricing of distribution network operators output by the upper-level model, with the electricity consumption strategies of greenhouse planting industry and livestock breeding industry as decision variables and the maximization of user benefits in agricultural parks as the optimization goal.

5. A distribution network-agricultural park double-layer optimization method according to claim 4, characterized in that: The upper model is: F=C FG +C BG +C PV +C SL +C DR Where: F is the comprehensive operating cost of the distribution network operator; C FG , C BG , C PV , C SL , C DR are the peak-valley difference smoothing cost, power supply guarantee cost, abandoned light cost, network loss cost and demand response compensation cost of the distribution network operator; c FG N is the cost coefficient for smoothing the peak-to-valley difference of the distribution network; N is the total number of nodes; P t,i is the net load value of node i in period t, the load is defined as positive and the power output is defined as negative; c BG Guaranteed supply cost per unit of electricity; is the loss load required to be guaranteed for node j in time period t; c PV The cost of curtailing solar power per unit of electricity; is the predicted maximum output value of distributed photovoltaic j in time period t; is the output value of distributed photovoltaic j called by the distribution network operator in time period t; c SL is the network loss cost coefficient; r ij is the resistance value between lines ij; I ij,t is the square value of the current of line ij in time period t; L is the line set.

6. A distribution network-agricultural park double-layer optimization method according to claim 4, characterized in that: The underlying model is: Where: is the basic electricity price charged by the distribution network operator at time t; N f is the set of nodes where the agricultural park is located; is the predicted basic power consumption of greenhouse planting industry in agricultural park at time t; is the predicted basic power consumption of livestock breeding industry at time t; The demand response electricity price compensation strategy formulated by the distribution network operator at time t; The actual power consumption of greenhouse planting industry in the agricultural park at time t after the real-time demand response strategy; The actual electricity consumption of the livestock breeding industry at time t after the real-time demand response strategy.

7. A distribution network-agricultural park double-layer optimization method according to claim 4, characterized in that: The constraints of the distribution network-agricultural park two-layer optimization model are: Where: P ij is the active power flowing between nodes ij; Q ij is the reactive power flowing between nodes ij; R ij is the resistance value between the distribution network lines ij; X ij is the reactance value between the distribution network lines ij; P L i,max is the maximum value of active power of load node i; Q L i,max is the maximum value of reactive power of load node i; is the square of the maximum voltage of load node i in the distribution network; is the square of the minimum voltage of load node i in the distribution network; is the square of the maximum current between the distribution network lines ij; ω ij is the Boolean variable of the connection of the ij node; ω ij =1 means that node j is connected; M is a very large positive number.

8. A distribution network-agricultural park double-layer optimization method according to claim 1, characterized in that: The improved particle swarm algorithm in step S4 is: Step A, initializing the population; Step B, calculate the fitness value of each particle in the population; Step C, updating the inertia weight according to the adaptive linear descent method, and sorting the population particles according to the fitness value; Step D: adopt the example learning strategy for the non-optimal particles in the population, adopt the random reverse learning strategy for the optimal particles in the population in the early iteration, and adopt the learning strategy combining general reverse learning and adaptive mutation mechanism in the late iteration; Step E: Calculate new fitness values ​​in parallel and select the best particle and the global optimum; Step F: Determine whether the termination condition is met. If so, terminate the iteration; if not, return to step C.

9. A distribution network-agricultural park double-layer optimization method according to claim 8, characterized in that: The step S4 comprises: Step S41, input the decision variables of the upper model in the distribution network-agricultural park two-layer optimization model, set the particle swarm algorithm parameters, and set the number of iterations k=0; Step S42: Use particle swarm algorithm to randomly generate m groups of distribution network operator pricing, and pass the parameters to the lower model; Step S43: The lower model sets prices for distribution network operators, calls the cplex solver to calculate the optimal electricity consumption strategies for greenhouse planting and livestock breeding with the goal of maximizing benefits, retains the current benefits of agricultural park users and returns the electricity consumption strategies to the upper model; Step S44: The upper model calculates and retains the current distribution network operator's revenue and pricing strategy based on the user's greenhouse planting industry electricity consumption strategy and livestock breeding industry electricity consumption strategy; Step S45, recording the particle positions, optimal positions of the population, particle fitness and optimal fitness of the population of m groups of pricing; Step S46: If the number of iterations is less than the set number of iterations, the particle swarm algorithm is used to generate new m groups of distribution network operator pricing and the process jumps to step S43, with the number of iterations k=k+1; if the number of iterations is greater than or equal to the set number of iterations, step S47 is executed; Step S47, output the optimal pricing strategy of the distribution network operator, the optimal electricity consumption strategy of the agricultural park users, the income of the distribution network operator and the benefits of the agricultural park users.

10. A distribution network-agricultural park two-layer optimization system, characterized in that: A method for executing a distribution network-agricultural park double-layer optimization method as claimed in any one of claims 1 to 9, comprising: Data acquisition unit, used to acquire basic data, including: distribution network topology, distributed photovoltaic and park agricultural facility parameters; A basic model building unit, used to build a mathematical model of distributed photovoltaic power generation, electricity consumption in greenhouse planting and electricity consumption in livestock breeding based on the basic data; An optimization model building unit, used to build a distribution network-agricultural park double-layer optimization model based on the data model and considering the decision conflicts between distribution network operators and agricultural park users in terms of benefit improvement; The optimization model solving unit is used to solve the distribution network-agricultural park double-layer optimization model by using an improved particle swarm algorithm combined with a cplex solver to obtain the optimal distribution network operator pricing strategy and the optimal agricultural park user electricity consumption strategy.

Citation Information

Cited By

  • Intelligent design method, system and equipment for light film-energy storage mushroom greenhouse considering frequency stability of rural power distribution network

    CN120982352A

  • A mushroom shed intelligent design method, system and equipment considering frequency stability of rural distribution network

    CN120982352B