Regional energy planning method based on urban information gridding principle and improved GA

Abstract

The invention discloses a regional energy planning method based on an urban information gridding principle and an improved GA (Genetic Algorithm), and the method comprises the steps: defining a planned region through a gridding method, carrying out the global optimization of the region through the optimization of an improved genetic algorithm, and seeking an urban distributed energy planning method with the lowest comprehensive cost. According to the invention, based on a city information model base and city model information, coordinates of a planned area are positioned through gridding division; marking the electrical load on a gridded model base coordinate as a'known 'condition; an improved genetic algorithm is adopted, a power supply load type and a link between a power supply load and an electrical load are used as variables, an initial population is constructed, and the lowest comprehensive cost formed by the construction cost of the power supply load, the operation cost of the power supply load and the construction cost of a power supply line of the power supply load is used as an optimization target; and performing global optimization by using an improved genetic algorithm so as to realize an optimal planning effect.

Description

Technical Field

[0001] The present invention relates to the field of distributed energy planning in the process of urban development, and in particular to a regional distributed energy planning method based on the urban information gridding principle and an improved genetic algorithm. Background Art

[0002] Distributed energy resources (DERs, such as photovoltaics, wind power, biomass, and geothermal energy) have become a key global energy development direction, ushering in a broader space for development. However, the planning and layout of distributed energy resources in urban development has become a significant challenge, requiring comprehensive consideration of multiple optimization objectives, including technical, economic, and environmental considerations. Traditional urban planning methods (such as linear programming) struggle to handle high-dimensional, nonlinear problems. Consequently, a comprehensive and systematic approach to this problem is currently lacking. Empirical or trial-and-error approaches are often employed, resulting in low planning efficiency, duplicated construction due to extensive cross-cutting, a lack of planning basis, and high actual construction and operating costs. Summary of the Invention

[0003] The purpose of this invention is to provide a regional distributed energy planning method based on the principle of urban information gridding and an improved genetic algorithm. This method combines the basic urban information stored in the existing City Information Model (CIM) platform with the platform's powerful computing power, and utilizes machine learning and computer-aided design methods to achieve scientific planning results, thereby solving the problems encountered in the background technology. The technical solutions adopted by this invention are as follows:

[0004] A regional energy planning method based on the urban information gridding principle and improved GA is characterized in that the method comprises the following steps:

[0005] Step (1), gridding the area to be planned;

[0006] Step (2), determining the coordinates of the power load;

[0007] Step (3) uses a dual-objective improved genetic algorithm to obtain the type C and location coordinates of distributed energy power supply loads to achieve comprehensive optimization of construction and operation costs, including the following steps:

[0008] (3.1), randomly generate P initial matrices of distributed energy supply loads Construct a system with p distributed energy power supply loads, and the type and location of each power supply load are A sample of the initial matrix of power supply load, namely

[0009] (3.2), generate the initial power supply relationship through the “nearest traversal algorithm”;

[0010] (3.3), calculate the fitness value:

[0011]

[0012] in, For sample Pop id The construction cost of all types of power supply loads, For sample Pop id The operating costs of all types of power supply loads in For sample Pop id The construction cost of power supply lines for all types of power supply loads;

[0013] (3.4), using the “random crossover method” to cross the initial power supply relationship and power supply load node type;

[0014] (3.5), use the “random mutation method” to mutate the power supply load node type;

[0015] (3.6) Iterate steps (3.1) to (3.5) and output the calculation results.

[0016] On the basis of adopting the above technical solutions, the present invention may also adopt the following further technical solutions, or use these further technical solutions in combination:

[0017] Step (1) specifically includes:

[0018] The base of the urban information model for the planned area is gridded, with the bottom and leftmost ends of the model base serving as the x-axis and y-axis of the grid coordinate system, respectively. The city is divided into finer grid cells, with each cell collecting and processing information including energy distribution, geographic data, and infrastructure.

[0019] During the gridding process, a uniform unit length e is used as a grid length;

[0020] According to the base coordinates of the gridded urban information model, the coordinate matrix of the planned area is established as Matrix = [(x, y)] m×n , where x is the horizontal coordinate; y is the vertical coordinate, x,y∈R + , m is the number of rows of the matrix, n is the number of columns of the matrix; the m×n matrix is ​​a standard rectangle. In the process of gridding the area to be planned, it is ensured that the coordinate matrix covers the entire planned area.

[0021] In step (2), the coordinate positions of the power loads in the planned area are determined and marked in the coordinate matrix. in It is assumed that the electricity load is located at the center of the grid.

[0022] In step (3.1), the types of distributed energy power supply loads and the corresponding possible positions of these distributed energy power supply loads in the coordinate matrix Matrix are randomly generated to form the initial matrix of distributed energy power supply loads in Randomly generated, corresponding to the coordinate position The types of distributed energy power supply loads include photovoltaic power, wind power, biomass power, and geothermal power. If the randomly generated distributed energy power supply load is located outside the planning area, the initial matrix is ​​invalid and deleted; the specific steps are as follows:

[0023] 1) Randomly generate a positive integer p, where p is the number of distributed energy power supply loads;

[0024] 2) Determine a coordinate position of the p distributed energy power supply loads: Use a random generation method to generate the horizontal coordinates vertical axis But the coordinates do not exceed the planning area coordinate matrix Matrix=[(x,y)] m×n range, i.e.

[0025] 3) Determine the type of one of the p distributed energy power supply loads: Assume that the number of types of distributed energy power supply loads available for selection in a certain planning area is K, where K∈N, that is, Randomly generate an integer k∈[1,K] in the range [1,K], then the set of distributed energy power supply load types corresponding to k is The location in the figure is the type of power supply load at that location;

[0026] 4) Repeat the above steps 2) and 3) p times to generate p distributed energy initial power supply load matrices Then a distributed energy power supply load is formed, and the type and location of each power supply load are A sample of the initial matrix of power supply load, namely Where id is the population number;

[0027] Formula (1) is used as the constraint condition, that is, the total power supply power of the generated initial power supply load matrix is ​​greater than the total power load power of the planning area:

[0028]

[0029] The planning area includes all power load types; otherwise, the generated initial power supply load matrix is ​​invalid and deleted.

[0030] In step (3.2), the specific steps are as follows:

[0031] For a power supply load generated in step (3.1), its coordinates are Power supply load type is Traverse the coordinate system, that is: let i and j are integers starting from 0, i.e. i, j∈R; Indicates that the location has both power supply load and power consumption load;

[0032] like It means that the traversed position is the coordinate position of the power load in the planned area, and then it is determined whether to establish an initial power supply relationship: if the power supply power provided by the power supply load is greater than the power consumption of the power load, the two establish an initial power supply relationship; otherwise, the initial power supply relationship is not established, and the traversal of the power load coordinates stops;

[0033] Among them, formula (2) is used as the coordinate constraint condition:

[0034]

[0035] Indicates that when the power supply load coordinate traversal exceeds the coordinate matrix range of the planning area, the power supply load will stop traversing regardless of whether there are remaining power supply loads;

[0036] The power supply loads that have stopped traversing and the power loads that have established an initial power supply relationship with a power supply load are marked, and no longer participate in the traversal process of other power supply loads, and do not repeatedly establish an initial power supply relationship with other power supply loads.

[0037] In step (3.3), the sample Pop id After all nodes have established random power supply relationships, calculate the fitness value Fitness; set the screening ratio ρ fit , according to the screening ratio ρ fit Discard samples with smaller fitness values ​​after calculation and replace them with samples with higher fitness values.

[0038] Step (3.4) specifically includes:

[0039] 1) Set a crossover probability ρ cro ;

[0040] 2) Crossing the initial power supply relationship; for the initial power supply relationship generated in step (3.2), according to the cross probability ρ cro Randomly select two pairs of power supply relationships and cross-swap them. If the "power constraint" condition is still met after cross-swap according to step (3.2), use the cross-swap sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fitJudge the fitness value, if its fitness is less than ρ fit If the crossover is successful, the crossover result is discarded; otherwise, the crossover result is retained and replaced with the original initial power supply relationship;

[0041] 3) Cross-powered load node type; according to the cross probability ρ cro Randomly select two power supply nodes and exchange their positions. If the "power constraint" condition is still met after the crossover according to step (3.2), use the crossover sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fit Determine the fitness value. If its fitness is less than ρ fit If the crossover is not found, the crossover result is retained and the original power supply node position is replaced.

[0042] Step (3.5) specifically includes:

[0043] 1) Set the mutation probability ρ mut ;

[0044] 2) From the sample matrix Pop id According to the mutation probability ρ mut Randomly select a power supply node and change its power supply load type; if the "power constraint" condition is still met after the mutation according to step (3.2), use the mutated sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fit Judge the fitness value; if its fitness is less than ρ fit If the crossover is not found, the mutation result is retained and the original power supply node position is replaced.

[0045] In step (3.6), set the number of iterations T. When the number of iterations T is reached, select the sample Pop with the highest fitness value. id , and its corresponding sample type and coordinate position are the results of the optimization plan.

[0046] The construction cost of the power supply load of the fitness value Fitness in step (3.3), the operating cost of the power supply load, and the power supply line construction cost of the power supply load are specifically described as follows:

[0047] (1) Construction cost of power supply load, including equipment purchase cost C pur and equipment installation cost C ins These two costs are mainly affected by the type of power supply equipment and are one-time fixed costs, namely:

[0048]

[0049] (2) The operating cost of the power supply load is composed of equipment operation and maintenance costs, equipment failure repair costs and equipment risk loss costs, namely:

[0050]

[0051] Where t is the equipment operating life; g(t) is the equipment operation and maintenance cost correction function, which is fitted by an n-order function; β and η are Weibull distribution functions used to describe the equipment failure rate, c repair is the maintenance cost per time; r(t) is the equipment risk function, which is fitted using the “least squares method”, c loss The cost of a single loss.

[0052] N is the design service life of the equipment;

[0053] (3) The construction cost of the power supply line for the power supply load; the “approximate shortest distance method” is used to calculate the sample Pop id The construction cost of power supply lines for all types of power supply loads is:

[0054]

[0055] in, For sample Pop id The coordinates of any power supply load; The coordinates of any power load that establishes a power supply relationship with the power supply load; the actual value is revised using the δ distance correction factor; c link is the construction cost per unit construction distance.

[0056] The method of the present invention is based on an urban information model base and urban model information, and locates the coordinates of the planned area through grid division; the power load is marked as a "known" condition on the gridded model base coordinates; an improved genetic algorithm is used, and the power supply load type and the power supply load and power load link are used as variables to construct an initial population, and the "lowest comprehensive cost" composed of the construction cost of the power supply load, the operating cost of the power supply load, and the power supply line construction cost of the power supply load is taken as the optimization goal; in the process, the "different comprehensive costs" of different power supply load types are considered, and the power loads provided are also different, so as to optimize different combinations, and use the improved genetic algorithm to perform global optimization, so as to achieve the best planning effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is the algorithm flow chart of the present invention;

[0058] Figure 2 This is a schematic diagram of the proportion of various types of energy in a simulation embodiment of the present invention;

[0059] Figure 3 This is a schematic diagram of the matching of power load and power supply load in the simulation embodiment of the present invention;

[0060] Figure 4 Taking a simulation embodiment as an example, the cost comparison diagram of the method of the present invention and the empirical planning method is shown; wherein, the method in the figure is the method of the present invention. DETAILED DESCRIPTION

[0061] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0062] The present invention provides a regional distributed energy planning method based on the principle of urban information gridding and an improved genetic algorithm. It proposes a method for urban distributed energy planning that uses a gridding method to define the planned area and uses an improved genetic algorithm to optimize the area for global optimization, seeking the "lowest overall cost" urban distributed energy planning method. This method addresses the current problems of planning in related fields that mainly rely on empirical methods, resulting in low planning efficiency, duplicate construction due to multiple overlapping areas, lack of planning basis, and high actual construction and operation costs. The method is based on an urban information model base and urban model information, and locates the coordinates of the planned area through grid division. The power load is marked as a "known" condition on the gridded model base coordinates. An improved genetic algorithm is used to construct an initial population, taking the power supply load type and the power supply load link as variables. The optimization goal is to achieve the "lowest overall cost" composed of the power supply load construction cost, power supply load operation cost, and power supply line construction cost. The process considers the different "comprehensive costs" of different power supply load types and the different power loads provided, thereby optimizing different combinations. The improved genetic algorithm is used for global optimization to achieve the best planning effect. To illustrate the effect of the present invention, the method of the present invention is described in detail below:

[0063] Step (1): Grid the area to be planned. This includes the following sub-steps:

[0064] (1.1) The urban information model base of the planned area is gridded, with the bottom and leftmost ends of the model base serving as the x-axis and y-axis of the grid coordinate system, respectively. The city is divided into finer grid cells, each of which collects and processes various information, including energy distribution, geographic data, infrastructure, etc.

[0065] (1.2) Unification of grid length: During the gridding process, a uniform unit length e is used as a grid length, which can be selected according to the specific scope of the planning area.

[0066] (1.3) Based on the gridded urban information model base coordinates (hereinafter referred to as coordinates), the coordinate matrix of the planned area is established as Matrix = [(x, y)] m×n , where x is the horizontal coordinate; y is the vertical coordinate, x,y∈R + .

[0067] According to the properties of the matrix, an m×n matrix must be a standard rectangle, while the planned area is usually a non-standard shape. Therefore, in the process of gridding the area to be planned, it is necessary to ensure that the coordinate matrix covers the entire planned area. Even if the edge of the planned area may not be filled with a grid due to the irregular outline of the planned area, a grid is still allocated to it.

[0068] Step (2): Determine the coordinates of the power load according to the actual situation. After gridding, the coordinate positions of the power load in the planned area can be determined and marked in the coordinate matrix. in For ease of calculation, it is assumed that these loads are located at the center of the grid where they are located.

[0069] Step (3): Use the dual-objective improved genetic algorithm to obtain the type C and location coordinates of the distributed energy power supply load to achieve comprehensive optimization of construction and operation costs. Specifically, it includes the following sub-steps:

[0070] (3.1) Randomly generate the types of distributed energy power supply loads and the corresponding possible positions of these distributed energy power supply loads in the coordinate matrix Matrix to form the initial matrix of distributed energy power supply loads in Randomly generated, corresponding to the coordinate position The types of distributed energy power supply loads include but are not limited to photovoltaic power, wind power, biomass power, and geothermal power. If the randomly generated distributed energy power supply load is located in the part that exceeds the actual planning range in step (1.3), the initial matrix is ​​invalid and deleted. The specific steps are as follows:

[0071] 1) Randomly generate a positive integer p, where p is the number of distributed energy supply loads.

[0072] 2) Determine a coordinate position of the p distributed energy power supply loads: Use a random generation method to generate the horizontal coordinates vertical axis But the coordinates should not exceed the planning area coordinate matrix Matrix=[(x,y)] m×n range, i.e.

[0073] 3) Determine the type of one of the p distributed energy power supply loads: Assume that the number of types of distributed energy power supply loads available for selection in a certain planning area is K, where K∈N, that is, Randomly generate an integer k∈[1,K] in the range [1,K], then the set of distributed energy power supply load types corresponding to k is The location in the figure is the type of load supplied at that location.

[0074] 4) Repeat the above steps 2) and 3) p times to generate p initial matrices of distributed energy power supply loads Then a distributed energy power supply load is formed, and the type and location of each power supply load are A sample of the initial matrix of power supply load, namely Where id is the population number.

[0075] 5) The constraint condition of step (3.1) is that the total power supply of the generated initial power supply load matrix is ​​greater than the total power load power of the planning area, that is:

[0076]

[0077] The planning area includes all power load types; otherwise, the generated initial power supply load matrix is ​​invalid and deleted.

[0078] In the actual programming process,

[0079] (3.2) Use the “nearest traversal algorithm” to generate the initial power supply relationship. The specific steps are as follows:

[0080] 1) For a power supply load generated in step (3.1), its coordinates are Power supply load type is Traverse the coordinate system, that is: let i and j are integers starting from 0, i.e. i, j∈R, Indicates that this location has both power supply load and power consumption load.

[0081] like It means that the traversed position is the coordinate position of the power load in the planned area, and then it is determined whether the initial power supply relationship is established: if the power supply power that the power supply load (remaining) can provide is greater than the power consumption of the power load, then the two establish an initial power supply relationship; otherwise, the initial power supply relationship is not established, and the traversal of the power supply load coordinates stops.

[0082] 2) Coordinate constraints: When

[0083]

[0084] Indicates that the power supply load coordinate traversal has exceeded the coordinate matrix range of the planning area, and the power supply load traversal will be stopped regardless of whether there are remaining power supply loads.

[0085] 3) The power supply loads that have stopped traversing and the power loads that have established an initial power supply relationship with a power supply load are marked, and they no longer participate in the traversal process of other power supply loads and do not repeatedly establish an initial power supply relationship with other power supply loads.

[0086] (3.3) Sample Pop id After all nodes (power supply load and power consumption load) have established random power supply relationships, the fitness value Fitness is calculated. The fitness value adopts the construction, operation cost and minimum function, that is:

[0087]

[0088] in, For sample Pop id The construction cost of all types of power supply loads, For sample Pop id The operating costs of all types of power supply loads in Sample Pop id The power supply line construction cost of all types of power supply loads in the system. Set the screening ratio to 0<ρ fit <5% (the ratio is adjustable), the samples with smaller fitness values ​​after calculation are discarded according to the screening ratio and replaced with samples with higher fitness values.

[0089] In the planning process, cost is usually an important factor to consider. Therefore, the cost of the power supply load of the fitness value Fitness is selected as the comprehensive minimum value of the construction cost, the operating cost of the power supply load, and the power supply line construction cost of the power supply load. The specific description is as follows:

[0090] (1) Construction cost of power supply load. Construction cost mainly includes equipment purchase cost C pur and equipment installation cost C ins These two costs are mainly affected by the type of power supply equipment and are one-time fixed costs.

[0091]

[0092] (2) Operating cost of power supply load. Although the operating cost is the same for different equipment, it is basically composed of equipment operation and maintenance cost, equipment failure repair cost and equipment risk loss cost, namely:

[0093]

[0094] Where t is the equipment operating life; g(t) is the equipment operation and maintenance cost correction function, which can be fitted by an n-order function; its overall law usually presents an exponential characteristic, that is, the operation and maintenance cost will gradually increase with the increase of time; β and η are Weibull distribution functions used to describe the equipment failure rate, c repair is the cost of a single repair; r(t) is the equipment risk function, which is usually provided by the manufacturer and can be fitted using the “least squares method”. loss is the cost of a single loss; N is the design service life of the equipment.

[0095] (3) The construction cost of the power supply line for the power supply load. The “approximate shortest distance method” is used to calculate the sample Pop id The construction cost of power supply lines for all types of power supply loads is:

[0096]

[0097] in, For sample Pop id The coordinates of any power supply load; The coordinates of all power loads that have established a power supply relationship with the power supply load. Considering that in actual construction projects, it is impossible to use the "shortest distance" for all line laying due to objective factors such as geographical terrain, the δ distance correction factor is used to revise the actual value, c link is the construction cost per unit construction distance.

[0098] In the actual planning process, in addition to cost factors, environmental factors, social benefits and other factors are usually factors that need to be considered. Therefore, in the actual application of the method of the present invention, on the one hand, other factors can be "costed" and integrated into the fitness function for global optimization; on the other hand, the above factors can be "normalized" and unified into a new, more inclusive and wider fitness function for optimization. Necessary adjustments based on actual conditions will not affect the effectiveness of the algorithm, and modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be deemed to be included in the scope of protection of the present invention.

[0099] (3.4) Use the “random cross method” to cross the initial power supply relationship and power supply load node type.

[0100] 1) Set a crossover probability 0<ρ cro <1.

[0101] 2) Cross initial power supply relationship. For the initial power supply relationship generated in step (3.2), according to the probability ρcro Randomly select two pairs of power supply relationships and cross-swap them; if the "power constraint" condition is still met after cross-swap according to step (3.2), use the cross-swap sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fit Judge the fitness value; if its fitness is less than ρ fit If the crossover is not successful, the crossover result is retained and replaces the original initial power supply relationship.

[0102] 3) Cross-power supply load node type. According to probability ρ cro Randomly select two power supply nodes and exchange their positions; if the "power constraint" condition is still met after the crossover according to step (3.2), use the crossover sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fit Judge the fitness value; if its fitness is less than ρ fit If the crossover is not found, the crossover result is retained and the original power supply node position is replaced.

[0103] (3.5) Use the “random variation method” to vary the power supply load node type.

[0104] 1) Set the mutation probability 0<ρ mut <<1.

[0105] 2) From the sample matrix Pop id The probability ρ mut Randomly select a power supply node and change its power supply load type. If the "power constraint" condition is still met after the mutation according to step (3.2), use the mutated sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fit Determine the fitness value. If its fitness is less than ρ fit If the crossover is not found, the mutation result is retained and the original power supply node position is replaced.

[0106] (3.6) Iterate the calculation steps (3.1) to (3.5) and output the calculation results. Set the number of iterations T; when the number of iterations T is reached, select the sample Pop with the highest fitness value. id , and its corresponding sample type and coordinate position are the results of the optimization plan.

[0107] Experimental and simulation results:

[0108] To verify the effectiveness of the method of the present invention, a simulation verification of the method of the present invention was carried out: assuming that the planned area can be gridded into a 50*50 coordinate system, there are 200 points suitable for the deployment of a distributed energy power supply system; there are 20 power loads in total, and to demonstrate the effectiveness of the method, the locations of the 20 power loads and their power consumption are randomly generated each time. The types of distributed energy available for planning include photovoltaic, wind power, biomass power generation, and geothermal power generation, with four types to choose from. The power loads that can be provided by each type of distributed power generation load are also randomly generated each time. The number of iterations of the genetic algorithm is 100,000. The simulation is based on MATLAB 2024b (64-bit) software. Figure 1 Shown is the calculation flow chart of the planning method.

[0109] Figure 2 The figure shows the combination relationship of various types of power generation loads in 10 simulation results. Figure 3 The corresponding relationship between the total power consumption and the total power generation in each simulation can be seen from the figure. Under the premise of ensuring that the total power generation is greater than the total power consumption and there is no power shortage, there is a good match between the power generation and the power consumption, and the overall load matching is between 85% and 94% (such as Figure 3 This shows that the power supply load type planned by the method of the present invention can better meet the power consumption requirements of the planned area while avoiding load waste.

[0110] at the same time, Figure 4 The following is a comparison curve of the total cost of using the traditional (random plus manual correction) planning method and the planning method using this method under the conditions of the total amount of electricity load generated in each simulation. Figure 4 It can be clearly seen that the overall cost (the entire planning area) optimized by the method of the present invention is more effective than that of the traditional planning method.

[0111] The above results demonstrate the excellent effectiveness of the proposed method in regional distributed energy planning. It should be noted that those skilled in the art will readily appreciate that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A regional energy planning method based on the principle of urban information gridding and improved GA is characterized by: The method comprises the following steps: Step (1), gridding the area to be planned; Step (2), determining the coordinates of the power load; Step (3) uses a dual-objective improved genetic algorithm to obtain the type C and location coordinates of distributed energy power supply loads to achieve comprehensive optimization of construction and operation costs, including the following steps: (3.1), randomly generate P initial matrices of distributed energy supply loads Construct a system with p distributed energy power supply loads, and the type and location of each power supply load are A sample of the initial matrix of power supply load, namely (3.2) Generate the initial power supply relationship through the "nearest traversal algorithm"; (3.3), calculate the fitness value: in, For sample Pop id The construction cost of all types of power supply loads, For sample Pop id The operating costs of all types of power supply loads in For sample Pop id The construction cost of power supply lines for all types of power supply loads; (3.4), using the "random cross method" to cross the initial power supply relationship and power supply load node type; (3.5) Use the "random mutation method" to mutate the power supply load node type; (3.6) Iterate steps (3.1) to (3.5) and output the calculation results.

2. The regional energy planning method based on the urban information gridding principle and improved GA according to claim 1 is characterized in that: Step (1) specifically includes: The base of the urban information model for the planned area is gridded, with the bottom and leftmost ends of the model base serving as the x-axis and y-axis of the grid coordinate system, respectively. The city is divided into finer grid cells, with each cell collecting and processing information including energy distribution, geographic data, and infrastructure. During the gridding process, a uniform unit length e is used as a grid length; According to the base coordinates of the gridded urban information model, the coordinate matrix of the planned area is established as Matrix = [(x, y)] m×n , where x is the horizontal coordinate; y is the vertical coordinate, x,y∈R + , m is the number of rows of the matrix, n is the number of columns of the matrix; the m×n matrix is ​​a standard rectangle. In the process of gridding the area to be planned, it is ensured that the coordinate matrix covers the entire planned area.

3. The regional energy planning method based on the urban information gridding principle and improved GA according to claim 2 is characterized in that: In step (2), Determine the coordinates of the power loads in the planned area and mark them in the coordinate matrix in It is assumed that the electricity load is located at the center of the grid.

4. The regional energy planning method based on the urban information gridding principle and improved GA according to claim 3 is characterized in that: In step (3.1), the types of distributed energy power supply loads and the corresponding possible positions of these distributed energy power supply loads in the coordinate matrix Matrix are randomly generated to form the initial matrix of distributed energy power supply loads in Randomly generated, corresponding to the coordinate position The types of distributed energy power supply loads include photovoltaic power, wind power, biomass power, and geothermal power. If the randomly generated distributed energy power supply load is located outside the planning area, the initial matrix is ​​invalid and deleted; the specific steps are as follows: 1) Randomly generate a positive integer p, where p is the number of distributed energy supply loads; 2) Determine a coordinate position of the p distributed energy power supply loads: Use a random generation method to generate the horizontal coordinates vertical axis But the coordinates do not exceed the planning area coordinate matrix Matrix=[(x,y)] m×n range, i.e. 3) Determine the type of one of the p distributed energy power supply loads: Assume that the number of types of distributed energy power supply loads available for selection in a certain planning area is K, where K∈N, that is, Randomly generate an integer k∈[1,K] in the range [1,K], then the set of distributed energy power supply load types corresponding to k is The location in the figure is the type of power supply load at that location; 4) Repeat the above steps 2) and 3) p times to generate p distributed energy initial power supply load matrices Then a distributed energy power supply load is formed, and the type and location of each power supply load are A sample of the initial matrix of power supply load, namely Where id is the population number; Formula (1) is used as the constraint condition, that is, the total power supply power of the generated initial power supply load matrix is ​​greater than the total power load power of the planning area: The planning area includes all power load types; otherwise, the generated initial power load matrix is ​​invalid and deleted.

5. The regional energy planning method based on the urban information gridding principle and improved GA according to claim 4 is characterized in that: In step (3.2), the specific steps are as follows: For a power supply load generated in step (3.1), its coordinates are Power supply load type is Traverse the coordinate system, that is: let i and j are integers starting from 0, that is, Indicates that the location has both power supply load and power consumption load; like It means that the traversed position is the coordinate position of the power load in the planned area, and then it is determined whether to establish an initial power supply relationship: if the power supply power provided by the power supply load is greater than the power consumption of the power load, the two establish an initial power supply relationship; otherwise, the initial power supply relationship is not established, and the traversal of the power load coordinates stops; Among them, formula (2) is used as the coordinate constraint condition: Indicates that when the power supply load coordinate traversal exceeds the coordinate matrix range of the planning area, the power supply load will stop traversing regardless of whether there are remaining power supply loads; The power supply loads that have stopped traversing and the power loads that have established an initial power supply relationship with a power supply load are marked, and no longer participate in the traversal process of other power supply loads, and do not repeatedly establish an initial power supply relationship with other power supply loads.

6. The regional energy planning method based on the urban information gridding principle and improved GA according to claim 5 is characterized in that: In step (3.3), the sample Pop id After all nodes have established random power supply relationships, calculate the fitness value Fitness; Set the screening ratio ρ fit , according to the screening ratio ρ fit Discard samples with smaller fitness values ​​after calculation and replace them with samples with higher fitness values.

7. The regional energy planning method based on the urban information gridding principle and improved GA according to claim 6 is characterized in that: Step (3.4) specifically includes: 1) Set a crossover probability ρ cro ; 2) Crossing the initial power supply relationship; for the initial power supply relationship generated in step (3.2), according to the cross probability ρ cro Randomly select two pairs of power supply relationships and cross-swap them. If the "power constraint" condition is still met after cross-swap according to step (3.2), use the cross-swap sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fit Judge the fitness value, if its fitness is less than ρ fit If the crossover is successful, the crossover result is discarded; otherwise, the crossover result is retained and replaced with the original initial power supply relationship; 3) Cross-powered load node type; according to the cross probability ρ cro Randomly select two power supply nodes and exchange their positions. If the "power constraint" condition is still met after the crossover according to step (3.2), use the crossover sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fit Determine the fitness value. If its fitness is less than ρ fit If the crossover is not found, the crossover result is retained and the original power supply node position is replaced.

8. The regional energy planning method based on the urban information gridding principle and improved GA according to claim 7 is characterized in that: Step (3.5) specifically includes: 1) Set the mutation probability ρ mut ; 2) From the sample matrix Pop id According to the mutation probability ρ mut Randomly select a power supply node and change its power supply load type; if the "power constraint" condition is still met after the mutation according to step (3.2), use the mutated sample Pop id Calculate the fitness value Fitness of all nodes and filter them according to the ratio ρ fit Judge the fitness value; if its fitness is less than ρ fit If the crossover is not found, the mutation result is retained and the original power supply node position is replaced.

9. The regional energy planning method based on the urban information gridding principle and improved GA as claimed in claim 1, characterized in that: Set the number of iterations T. When the number of iterations T is reached, select the sample Pop with the highest fitness value. id , and its corresponding sample type and coordinate position are the results of the optimization plan.

10. The regional energy planning method based on the urban information grid principle and improved GA according to claim 1, characterized in that: The construction cost of the power supply load of the fitness value Fitness in step (3.3), the operating cost of the power supply load, and the power supply line construction cost of the power supply load are specifically described as follows: (1) Construction cost of power supply load, including equipment purchase cost C pur and equipment installation cost C ins These two costs are mainly affected by the type of power supply equipment and are one-time fixed costs, namely: (2) The operating cost of the power supply load is composed of equipment operation and maintenance costs, equipment failure repair costs and equipment risk loss costs, namely: Where t is the equipment operating life; g(t) is the equipment operation and maintenance cost correction function, which is fitted by an n-order function; β and η are Weibull distribution functions used to describe the equipment failure rate, c repair is the maintenance cost per time; r(t) is the equipment risk function, which is fitted using the "least squares method", c loss is the cost of a single loss. N is the design service life of the equipment; (3) The construction cost of the power supply line for the power supply load; the "approximate shortest distance method" is used to calculate the sample Pop id The construction cost of power supply lines for all types of power supply loads is: in, For sample Pop id The coordinates of any power supply load; The coordinates of any power load that establishes a power supply relationship with the power supply load; the actual value is revised using the δ distance correction factor; c link is the construction cost per unit construction distance.

Images