Comprehensive energy system station network multi-layer collaborative optimization method based on GIS three-dimensional clustering

Abstract

The invention relates to an integrated energy system station network multilayer collaborative optimization method based on GIS three-dimensional clustering, and the method comprises the following steps: 1, carrying out the equipment configuration and cost calculation of a three-dimensional clustering site selection result; 2, abstracting a researched area, and establishing a pipe network layout planning cost model; 3, solving according to a pipe network layout plan; 4, establishing a regional distributed energy system station network collaborative planning target model, and performing regional distributed energy system station network collaborative planning; and 5, solving according to the station network collaborative planning of the regional distributed energy system. The method provided by the invention is clear in logic and high in operability, can be suitable for locating and sizing planning of the regional comprehensive energy station, and avoids the problems of planning splitting, data dimension limitation, insufficient pipe network economy and the like of a traditional method.

Description

Technical Field

[0001] This invention relates to an optimization method, specifically a multi-layer collaborative optimization method for integrated energy system stations and networks based on GIS three-dimensional clustering. This method is logically clear, highly operable, and applicable to the site selection and capacity planning of regional integrated energy stations, avoiding problems such as planning fragmentation, data dimension limitations, and insufficient pipeline economics in traditional methods. Background Technology

[0002] With the goals of "peak carbon emissions by 2030 and carbon neutrality by 2060" being set, energy systems are transforming towards cleaner, more efficient, and integrated systems. Integrated energy systems, through the synergistic complementarity and cascaded utilization of multiple energy sources such as electricity, heat, cooling, and gas, have become an important way to improve energy efficiency and reduce carbon emissions. As an important component of this system, integrated energy stations undertake key functions of energy conversion, storage, and distribution, and the rationality of their site selection and capacity configuration directly affects the overall performance of the system.

[0003] However, in the context of energy transition, the high proportion of renewable energy integration has highlighted grid stability issues, and existing site selection and capacity determination studies generally have limitations: traditional methods optimize energy station site selection, equipment capacity determination, and pipeline layout step by step, ignoring the coupling relationship among the three (e.g., site selection affects pipeline costs, and pipeline layout restricts equipment selection). Existing clustering site selection only considers geographical coordinates (two-dimensional) and does not integrate energy consumption data, resulting in a mismatch between site layout and energy demand. The economic efficiency of the pipeline network is insufficient, and pipeline optimization relies on static weights (such as distance), without dynamically coupling pipe diameter selection and flow distribution.

[0004] The existing integrated energy station planning methods still have the following key issues that urgently need to be addressed:

[0005] First, the planning is fragmented. Traditional methods optimize energy station site selection, equipment capacity determination, and pipeline layout in separate steps, ignoring the coupling relationship among the three (such as site selection affecting pipeline costs, and pipeline layout restricting equipment selection). Energy station site selection and pipeline planning are optimized sequentially, and the objective function is not unified.

[0006] Second, there is a limitation in data dimensions. Existing clustering site selection only considers geographical coordinates (two-dimensional) and does not integrate energy consumption data, resulting in a mismatch between site layout and energy demand. The clustering algorithm has a single input dimension (only coordinates) and does not incorporate load data.

[0007] Third, the economic efficiency of the pipeline network is insufficient. Pipeline network optimization relies on static weights (such as distance) and does not dynamically couple pipe diameter selection with flow distribution. The impact of energy station site selection is not considered during pipeline network planning. Dijkstra's algorithm has inherent defects, is sensitive to the calculation order, and cannot handle pipeline sharing. Summary of the Invention

[0008] To address the aforementioned problems, the main objective of this invention is to provide a multi-layer collaborative optimization method for integrated energy system stations based on GIS three-dimensional clustering. This method is logically clear, highly operable, and applicable to the site selection and capacity planning of regional integrated energy stations, avoiding the problems of planning fragmentation, data dimension limitations, and insufficient pipeline economics of traditional methods.

[0009] The present invention solves the above-mentioned technical problems through the following technical solution: a multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering, wherein the multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering includes the following steps:

[0010] Step 1: Use K-means clustering to select the site for the integrated energy station based on the geographical location and building energy consumption data processed by AcrGIS. Combine the annual time-of-use data of cooling, heating and power load of the integrated energy station to perform equipment configuration and cost calculation based on the three-dimensional clustering site selection results.

[0011] The second step is to abstract the area under study and establish a pipeline layout planning cost model; solve for the optimal pipeline layout that maximizes the economic efficiency of the pipeline and satisfies the corresponding constraints, and establish a pipeline layout planning cost model.

[0012] Step 3: Solve the problem based on the pipeline layout plan;

[0013] Step 4: Conduct coordinated optimization of energy station site selection, pipeline layout, and equipment capacity determination; establish a regional distributed energy system station-network coordinated planning target model; and carry out regional distributed energy system station-network coordinated planning.

[0014] Step 5: Solve the problem based on the regional distributed energy system station-network collaborative planning.

[0015] In a specific implementation of this invention, the first step includes the following steps: using ArcGIS to classify building usage types and obtain the building's plot area, center coordinates, and building energy consumption data; performing K-means clustering analysis on the target data, compiling the optimal k-value candidate range, and using the elbow method to determine the optimal k-value within the range that meets the clustering requirements; obtaining the final clustering result after screening, and substituting the centroid coordinates into ArcMap to obtain the site selection of the integrated energy station under three-dimensional clustering; performing time-sharing prediction of regional cooling, heating, and electricity loads based on the multi-factor method, and calculating the energy consumption intensity and spatial distribution pattern of each land use type by combining historical data and building type characteristics, thereby obtaining the time-sharing data of the integrated energy station's annual cooling, heating, and electricity loads; and building an equipment configuration and economic analysis model to perform equipment configuration and cost calculation on the three-dimensional clustering site selection results.

[0016] In a specific implementation of the present invention, the second step includes the following steps: the objective function for optimizing the pipeline layout includes four parts, namely the construction cost of the pipeline system, the annual operating cost of the circulating water pump, the energy loss cost of the pipeline system, and the depreciation and maintenance cost of the pipeline system.

[0017] In a specific embodiment of the present invention, the second step is specifically as follows:

[0018]

[0019] The construction cost of a pipeline system is related to the pipe diameter and length. For the constructed undirected graph G(V,E,W), this cost can be expressed by the following formula:

[0020]

[0021] The pipelines in this section are constructed using direct burial. There are 13 different pipe diameters: DN200, DN250, DN300, DN350, DN400, DN450, DN500, DN600, DN700, DN800, DN900, DN1000, and DN1200.

[0022] The operating cost of a circulating water pump is related to the pump's design flow rate, electromechanical efficiency, operating pressure, and electricity price, and is calculated using the following formula:

[0023]

[0024] Energy loss costs in a pipeline system are directly related to the pipe diameter, length, and insulation thickness. The energy loss of a pipeline system is expressed by the following formula:

[0025]

[0026] Depreciation and maintenance costs of pipeline systems: In the actual operation of pipeline systems, there are various foreseeable and unforeseeable losses.

[0027] In specific embodiments of this invention, the cost of this part must be considered when calculating the economics of the pipeline network; this cost is highly uncertain, and in practice, engineering designers often calculate the cost according to a fixed ratio, calculated by the ratio of equipment depreciation rate and maintenance cost, specifically expressed as:

[0028]

[0029] In the design of regional distributed energy system pipeline networks, the selection of pipe diameter is influenced by multiple factors, including flow velocity and flow rate. Both flow velocity and flow rate have certain limitations, as detailed below:

[0030]

[0031] 0.5 <v ij <3.5 (7)

[0032] 200≤d ij ≤1200 (8).

[0033] In a specific embodiment of the present invention, the third step includes the following steps:

[0034] (301) Construct an undirected graph G(V,E,W) based on the geographic information of the planning area;

[0035] (302) Input road network information, simulation settings parameters, and energy station location information;

[0036] (303) Based on the interrelationship of each vertex in network G, solve for the edge matrix E. The specific calculation method is as follows;

[0037]

[0038] (304) Input the genetic algorithm control parameters and encode the parameters to optimize the calculation order of energy-consuming buildings;

[0039] (305) Randomly generate an initial population, set Gen=1, and obtain the Gen generation parent population sample under this layout;

[0040] (306) Calculate the fitness of the parent population in the Gen generation;

[0041] (307) Based on this fitness, selection, crossover and mutation are performed to generate a new generation of population, Gen = Gen+1;

[0042] (308) Repeat steps (305)-(307) until convergence is achieved;

[0043] (309) Save the optimized calculation order of energy-consuming buildings;

[0044] (310) Initialize the full adjacency matrix W of the pipeline cost, which is calculated from the previously input parameters as follows:

[0045]

[0046] (311) The path of the i-th energy-consuming building is calculated using Dijkstra's algorithm. The path from the energy-consuming building to all energy stations and the increased cost are calculated. The energy station with the lowest cost is selected according to the greedy strategy. Let i = i + 1.

[0047] (312) Update the cost weight adjacency matrix of the pipeline network. The weight is the increase in pipeline network cost caused by the new access load point. Consider the corresponding economically optimal pipe diameter according to the flow of different pipe segments. The calculation formula is as follows.

[0048]

[0049] (313) Repeat steps (11)-(12) until all energy-consuming buildings have been calculated and save the layout and pipe diameter selection of the entire pipeline system;

[0050] (314) Calculate the objective function C pl ;

[0051] (315) Calculate the basics, output the objective function value, pipeline layout and pipe diameter of each segment.

[0052] In a specific implementation of the present invention, the fourth step includes the following steps: the objective function of the mathematical model for the coordinated planning of regional distributed energy system stations and networks consists of two parts, namely the annualized cost of the energy station and the annualized cost of the pipeline network system.

[0053] In a specific embodiment of the present invention, the objective function calculation formula and the energy station cost calculation formula are as follows:

[0054] min C total =min(C es,2 +C pl (9)

[0055] The investment cost of an energy station is usually calculated based on the installed capacity of the equipment. The land cost and civil engineering cost of the energy station are usually estimated based on the power load of the energy station.

[0056] 9. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering as described in claim 8, characterized in that: the investment cost of various energy sources can be expressed by the following equation:

[0057]

[0058] Therefore, the total annualized cost of the energy station can be calculated using the following formula:

[0059]

[0060] In a specific embodiment of the present invention, the fifth step includes the following steps:

[0061] (501) Construct an undirected graph G(V,E,W) based on the geographic information of the planning area;

[0062] (502) Input the parameter information required for the calculation, including genetic algorithm control parameters, economic parameters, etc.;

[0063] (503) Solve for the edge matrix E based on the interrelationships between the vertices in network G;

[0064] (504) Determine the upper limit of the number of energy stations and the current number of energy stations;

[0065] (505) K-means clustering is used to initially determine the location of the energy station, and the location of the energy station is gradually optimized by combining the relative load distance and the energy supply range, and the optimized location of the energy station is saved.

[0066] (506) Genetic algorithm is used to encode and optimize the calculation order of energy-consuming buildings;

[0067] (507) Generate an initial population, set Gen=1, and obtain the Gen generation parent population sample. Calculate the fitness of this population.

[0068] (508) Based on the fitness of each individual in the Gen generation parent population, perform selection, crossover and mutation operations to generate the next generation population sample. Let Gen = Gen+1, and the Gen generation parent population sample is obtained.

[0069] (509) Determine whether the genetic algorithm has converged at this time. If it has not converged, execute step (506). If it has converged, execute step (510).

[0070] (510) The optimized energy-consuming building calculation order is obtained after calculation by genetic algorithm;

[0071] (511) Initialize the cost weight adjacency matrix W and the diameter matrix in network G. Calculate them according to the previous calculation results. The initial values ​​of the diameter matrix are all 0.

[0072] (512) The current calculation is for the i-th energy-consuming building. Dijkstra's algorithm is used to calculate the path from the energy-consuming building to all energy stations and the corresponding cost increment. The energy station with the lowest cost increment is selected according to the greedy strategy.

[0073] (513) Based on the selected energy station, record the corresponding path and the flow rate, velocity and pipe diameter of each pipe segment, and update the cost weight adjacency matrix and pipe diameter matrix. The cost value is the increase in network cost caused by the new energy-consuming building, and the pipe diameter is the most economical pipe diameter under this flow rate.

[0074] (514) Determine whether all energy-consuming buildings have been calculated. If not, proceed to step (512); otherwise, proceed to step (515).

[0075] (515) Save the obtained pipeline layout and pipe diameter of each pipe section, and calculate the corresponding annualized cost of the pipeline system.

[0076] (516) Determine the energy supply range of each energy station according to the pipeline network layout, calculate the cooling, heating and power supply capacity of each energy station, and calculate the annualized cost of the energy station.

[0077] (517) Determine whether the current number of energy stations has reached the upper limit. If not, proceed to step (504); otherwise, proceed to step (518).

[0078] (518) Output the optimal number of energy stations, and determine the location and energy supply range of the energy stations. Calculate the capacity of the energy station equipment based on the cooling, heating and power load requirements of the energy stations.

[0079] (519) The calculation ends, and the objective function value, number of energy stations, location of energy stations, pipeline layout, pipeline diameter and energy station equipment capacity are output.

[0080] Based on the above steps, the solution to the regional distributed energy system station-network coordinated optimization problem is finally obtained.

[0081] The positive and progressive effects of this invention are as follows: The multi-level collaborative optimization method for integrated energy system stations and networks based on GIS three-dimensional clustering provided by this invention has the following advantages: The method proposed in this invention has clear logic and strong operability, and can be applied to the site selection and capacity planning of regional integrated energy stations, avoiding the problems of planning fragmentation, data dimension limitation and insufficient network economy of traditional methods. Attached Figure Description

[0082] Figure 1 This is a schematic diagram of land use and hydrological data for the Shanghai area.

[0083] Figure 2 This is a schematic diagram showing the locations of Dazhi River and Jinhui Port.

[0084] Figure 3 This is a schematic diagram of the land use in the Lingang area according to the present invention.

[0085] Figure 4 This is a comprehensive distribution map of land use types in the Lingang area, as presented in this invention.

[0086] Figure 5 This is a comprehensive distribution map (latitude and longitude map) of land use types in the Lingang area in this invention.

[0087] Figure 6 This is a map showing the area and center coordinates of the land parcel in the attribute table of this invention.

[0088] Figure 7 Functional division map of the port area in the invention.

[0089] Figure 8 This is a schematic diagram of the final research area in this invention.

[0090] Figure 9-1 This is a coordinate graph generated for the k=16 three-dimensional clustering in this invention.

[0091] Figure 9-2 This is a coordinate graph generated for the three-dimensional clustering of k=17 in this invention.

[0092] Figure 10 This is a site selection diagram for a three-dimensional cluster-based integrated energy station in this invention.

[0093] Figure 11 This is a three-dimensional clustering grouping map in this invention.

[0094] Figure 12 This invention optimizes the site selection and pipeline layout of energy stations.

[0095] Figure 13 This is the predicted annual hydrogen demand of the construction industry in the Lingang 103 area in 2035, as presented in this invention.

[0096] Figure 14 This represents the typical daily hydrogen demand predicted in this invention.

[0097] Figure 15 This is a schematic diagram of the overall structure of the present invention. Detailed Implementation

[0098] The preferred embodiments of the present invention are given below with reference to the accompanying drawings to illustrate the technical solution of the present invention in detail.

[0099] Figure 15 This is a schematic diagram of the overall structure of the present invention, as shown below. Figure 15 As shown: This invention provides a multi-layer collaborative optimization method for integrated energy system station networks based on GIS three-dimensional clustering. The core objectives include:

[0100] Significantly improving planning coordination and economy, the system systematically addresses the pain point of insufficient coordination between "station-grid-load" in integrated energy system planning through four major innovations: three-dimensional data fusion, dynamic weight matrix, closed-loop coordination mechanism, and algorithm fusion.

[0101] To optimize the accuracy of energy station site selection, 3D data was integrated to improve demand matching. Energy consumption was introduced as a third dimension (coordinates x, y + energy consumption value) into the GIS platform, and clustering was performed using an improved K-means algorithm (elbow method to determine the k value). Implementation cases demonstrate that 3D clustering leads to a more rational site layout, a high degree of matching between centroid distribution and data density, balanced load across the energy station coverage area, avoidance of resource waste, and reduction of ineffective site investment.

[0102] Enhance the economic efficiency and dynamic adaptability of the pipeline network by dynamically coupling the pipeline cost model with pipe diameter selection and flow distribution to reduce redundant construction.

[0103] To achieve the present invention, the specific method employed is as follows:

[0104] First, K-means clustering was used to select the site for the integrated energy station based on the geographical location and building energy consumption data processed by AcrGIS. Combined with the time-of-use data of the integrated energy station's annual cooling, heating and power loads, the equipment configuration and cost calculation were performed on the three-dimensional clustering site selection results.

[0105] ArcGIS was used to classify building usage types and obtain the building's plot area, center coordinates, and building energy consumption data. K-means clustering analysis was performed on the target data. The optimal k-value candidate range was compiled, and the elbow method was used to determine the best k-value within the range that met the clustering requirements. After screening, the final clustering results were obtained. The centroid coordinates were substituted into ArcMap to obtain the site selection of the integrated energy station under three-dimensional clustering. Time-of-use forecasting of regional cooling, heating, and electricity loads was performed based on the multi-factor method. Combining historical data and building type characteristics, the energy intensity and spatial distribution patterns of each land use type were calculated, obtaining the annual time-of-use data for the integrated energy station's cooling, heating, and electricity loads. An equipment configuration and economic analysis model was built to calculate equipment configuration and costs based on the three-dimensional clustering site selection results.

[0106] Then, the area under study is abstracted, and a pipeline layout planning cost model is established; the optimal pipeline layout is solved to make the pipeline network economically optimal and meet the corresponding constraints, and a pipeline layout planning cost model is established.

[0107] The objective function for pipeline network layout optimization mainly consists of four parts: the construction cost of the pipeline system, the annual operating cost of the circulating water pumps, the energy loss cost of the pipeline system, and the depreciation and maintenance cost of the pipeline system. Specifically, it is expressed as follows:

[0108]

[0109] The construction cost of a pipeline system is mainly related to the pipe diameter and length. For the constructed undirected graph G(V,E,W), this cost can be expressed by the following formula:

[0110]

[0111] The pipelines in this section are constructed using direct burial. There are 13 different pipe diameters: DN200, DN250, DN300, DN350, DN400, DN450, DN500, DN600, DN700, DN800, DN900, DN1000, and DN1200.

[0112] The operating cost of a circulating water pump is mainly related to the pump's design flow rate, electromechanical efficiency, operating pressure, and electricity price, and can be calculated using the following formula:

[0113]

[0114] Energy loss costs in pipeline systems are directly related to factors such as pipe diameter, length, and insulation thickness. The energy loss in a pipeline system is expressed by the following formula:

[0115]

[0116] Depreciation and maintenance costs of pipeline systems: In the actual operation of pipeline systems, various foreseeable and unforeseeable losses exist. Therefore, this cost must be considered when calculating the economics of a pipeline system. This cost is highly uncertain, and in practice, engineering designers often calculate costs using fixed ratios, typically the ratio of equipment depreciation rate to maintenance cost, specifically expressed as:

[0117]

[0118] In the design of regional distributed energy system pipeline networks, the selection of pipe diameter is influenced by multiple factors, including flow velocity and flow rate. Both flow velocity and flow rate have certain limitations, as detailed below:

[0119]

[0120] 0.5 <v ij <3.5 (7)

[0121] 200≤d ij ≤1200 (8)

[0122] Secondly, the solution is derived based on the pipeline layout plan, and the process is as follows:

[0123] 1) Construct an undirected graph G(V,E,W) based on the geographic information of the planning area;

[0124] 2) Input relevant road network information, simulation settings, and energy station location information;

[0125] 3) Based on the interrelationships of the vertices in network G, solve for the edge matrix E. The specific calculation method is as follows;

[0126]

[0127] 4) Input the genetic algorithm control parameters, encode the parameters, and optimize the calculation order of energy-consuming buildings;

[0128] 5) Randomly generate an initial population, set Gen=1, and obtain the Gen-th generation parent population sample under this layout;

[0129] 6) Calculate the fitness of the parent population in the Gen generation;

[0130] 7) Based on this fitness, selection, crossover, and mutation are performed to generate a new generation of population, Gen = Gen+1;

[0131] 8) Repeat steps (5)-(7) until convergence is achieved;

[0132] 9) Save the optimized calculation order for energy-consuming buildings;

[0133] 10) Initialize the full adjacency matrix W of the pipeline cost, calculated from the previously input parameters, as follows:

[0134]

[0135] 11) Currently, the path of the i-th energy-consuming building is calculated. Dijkstra's algorithm is used to calculate the path from the energy-consuming building to all energy stations and the increased cost. The energy station with the lowest cost is selected according to the greedy strategy, and i = i + 1 is set.

[0136] 12) Update the cost weight adjacency matrix of the pipeline network. The weight is the increase in pipeline network cost caused by the new access load point. Consider the corresponding economically optimal pipe diameter according to the flow of different pipe segments. The calculation formula is as follows.

[0137]

[0138] 13) Repeat steps (11)-(12) until all energy-consuming buildings have been calculated, and save the layout and pipe diameter selection of the entire pipeline system;

[0139] 14) Calculate the objective function C pl ;

[0140] 15) Calculate the basics, output the objective function value, pipeline layout, and pipe diameter of each segment.

[0141] Next, we will conduct coordinated optimization of energy station site selection, pipeline layout, and equipment capacity determination, establish a regional distributed energy system station-network coordinated planning target model, and carry out regional distributed energy system station-network coordinated planning.

[0142] The objective function of the mathematical model for coordinated planning of regional distributed energy systems consists of two parts: the annualized cost of the energy stations and the annualized cost of the pipeline system. The formulas for calculating the objective function and the energy station cost are as follows:

[0143] min C total =min(C es,2 +C pl (9)

[0144] The investment cost of an energy station is typically calculated based on the installed capacity of the equipment. Land costs and civil engineering costs are usually estimated based on the station's electrical load. The investment costs of various energy sources can be expressed by the following equation:

[0145]

[0146]

[0147] Therefore, the total annualized cost of the energy station can be calculated using the following formula:

[0148]

[0149] Finally, the solution is obtained based on the regional distributed energy system network collaborative planning, and the process is as follows:

[0150] 1) Construct an undirected graph G(V,E,W) based on the geographic information of the planning area;

[0151] 2) Input the required parameters for the calculation, including genetic algorithm control parameters, economic parameters, etc.;

[0152] 3) Solve for the edge matrix E based on the interrelationships between the vertices in network G;

[0153] 4) Determine the upper limit for the number of energy stations; the current number of energy stations.

[0154] 5) K-means clustering is used to initially determine the location of energy stations, and the location of energy stations is gradually optimized by combining relative load distance and energy supply range, and the optimized location of energy stations is saved;

[0155] 6) Genetic algorithms are used to encode and optimize the calculation order of energy-consuming buildings;

[0156] 7) Generate an initial population, let Gen = 1, and obtain the parent population sample of the Gen generation. Calculate the fitness of this population.

[0157] 8) Based on the fitness of each individual in the Gen generation parent population, perform selection, crossover, and mutation operations to generate the next generation population sample. Let Gen = Gen+1, and the Gen generation parent population sample is obtained.

[0158] 9) Determine whether the genetic algorithm has converged at this time. If it has not converged, execute step (6). If it has converged, execute step (10).

[0159] 10) The optimized calculation order of energy-consuming buildings is obtained after calculation by the genetic algorithm;

[0160] 11) Initialize the cost weight adjacency matrix W and the diameter matrix in network G, based on the previous calculation results, where the initial values ​​of the diameter matrix are all 0;

[0161] 12) Currently, calculate the i-th energy-consuming building. Use Dijkstra's algorithm to calculate the path from the energy-consuming building to all energy stations and the corresponding cost increment. Select the energy station with the lowest cost increment according to the greedy strategy.

[0162] 13) Based on the selected energy station, record the corresponding path and the flow rate, velocity and pipe diameter of each pipe segment, and update the cost weight adjacency matrix and pipe diameter matrix. The cost value is the increase in network cost caused by the new energy-consuming building, and the pipe diameter is the most economical pipe diameter under this flow rate.

[0163] 14) Determine whether all energy-consuming buildings have been calculated. If not, proceed to step (12); otherwise, proceed to step (15).

[0164] 15) Save the obtained pipeline layout and pipe diameter of each segment, and calculate the corresponding annualized cost of the pipeline system;

[0165] 16) Determine the energy supply range of each energy station based on the pipeline network layout, calculate the cooling, heating and power supply capacity of each energy station, and calculate the annualized cost of the energy station.

[0166] 17) Determine whether the current number of energy stations has reached the upper limit. If not, proceed to step (4); otherwise, proceed to step (18).

[0167] 18) Output the optimal number of energy stations, and determine the location and energy supply range of the energy stations. Calculate the capacity of the energy station equipment based on the cooling, heating and power load requirements of the energy stations.

[0168] 19) The calculation is complete. Output the objective function value, number of energy stations, location of energy stations, pipeline layout, pipeline diameter, and energy station equipment capacity.

[0169] Based on the above steps, the solution to the regional distributed energy system station-network coordinated optimization problem is finally obtained.

[0170] The following is a specific embodiment of the present invention:

[0171] I. Map Import

[0172] (1) Data Acquisition

[0173] Find and download geographic information data for Shanghai from the Geofabrik website.

[0174] (2) Data import

[0175] (3) Import the land use and hydrological data into ArcMap to obtain data for Shanghai, such as... Figure 1 As shown.

[0176] (4) Preliminary site selection

[0177] According to the regional definition of Lingang, the Lingang area is mainly located in the southeastern part of Shanghai, south of the Dazhi River and east of the Jinhui Port, including Xiaoyangshan Island and the area south of Shanghai Pudong International Airport, with a total area of ​​873 square kilometers. By locating the Dazhi River and Jinhui Port in the hydrological layer attribute table and filtering out the area enclosed by the Dazhi River, Jinhui Port, and the East China Sea, the overall regional information of Lingang is obtained, such as... Figure 2 As shown.

[0178] Remove other data for the Shanghai area and hide the hydrological data for the Lingang area, retaining only the data for the Dazhi River, Jinhui Port, and land use in the Lingang area. Figure 3 As shown.

[0179] (5) Draw block diagram

[0180] According to the comprehensive attribute table, land use data includes: cemetery, commercial, farmland, forest, grassland, industrial, meadow, military, orchard, park, quarry, recreation ground, residential, retail, and scrub, totaling 15 categories. To visually illustrate the distribution of land use types, we have categorized different land uses on the map, such as... Figure 4 As shown.

[0181] (6) Draw the latitude and longitude grid

[0182] Add an edit grid to the layer, setting the x-axis and y-axis intervals to 2′. Then, add details such as a north arrow, scale bar, and legend, and optimize the image, such as... Figure 5 As shown.

[0183] II. Extracting Area and Coordinate Data

[0184] The attribute table of the imported land use SHP file shows that the current layer uses the geographic coordinate system GCS_WGS_1984. However, to obtain the plot area and center coordinates, it is necessary to generate them in a projected coordinate system.

[0185] Add a generated projected coordinate system WGS_1984_Web_Mercator_Auxiliary_Sphere to the original map layer, with the projection category being Mercator_Auxiliary_Sphere. Add fields "area", "CentralX", and "CentralY" to the attribute table. These fields are used to calculate the area data and center coordinates of each parcel, such as... Figure 6 As shown.

[0186] III. Determining the final research area

[0187] Depend on Figure 4 It can be seen that there are still many agricultural lands in the Lingang area, as well as many woodlands and lawns, and there are also many blank areas in the data that have not been collected.

[0188] according to Figure 7 As can be seen, the currently developed areas in Lingang are mainly concentrated in the core area of ​​Dishui Lake in the east, the comprehensive industrial area, the advanced intelligent area, and the emerging industrial park. In addition, Figure 4 Residential, commercial, and industrial clusters are also concentrated in the aforementioned areas. Therefore, combining the two maps and the research objectives, regions such as... Figure 8 As shown.

[0189] The land use types include: commercial, industrial, recreation ground, residential, and retail. Relevant parameters for the site selection area are shown in Table 1.

[0190] Table 1(a) Statistics on Land Use Type and Area

[0191]

[0192]

[0193] Table 1(b) Geographical Location Data (Selected Section)

[0194]

[0195] Different cities and regions have different standards for floor area ratios. According to the "Technical Regulations for Urban Planning Management in Shanghai (Land Use and Building Management)," the following floor area ratios can be obtained and the building area can be calculated. The results are shown in Table 2.

[0196] Table 2(a) Floor Area Ratio & Building Area

[0197]

[0198] Table 2(b) Specific Plot Ratios & Building Areas (Partial)

[0199]

[0200]

[0201] According to the "2023 Shanghai Municipal Government Office Buildings and Large Public Buildings Energy Consumption and Carbon Emission Monitoring and Analysis Report", the Unified Standard for Energy-Saving Design of Industrial Buildings GB 51245-2017, and the Urban Power Planning Code GB / T 50293-2014, the energy intensity of various types of buildings can be obtained through on-site visits, as shown in Table 3.

[0202] Table 3 Energy Intensity of Various Building Types

[0203]

[0204] The annual energy consumption of the building can be calculated, as shown in Table 4.

[0205] Table 4(a) Energy consumption of various types of buildings

[0206]

[0207] Table 4(b) Energy Consumption of Specific Buildings (Partial)

[0208]

[0209] After obtaining the energy consumption of each plot, the x, y values ​​and energy consumption can be used for three-dimensional clustering. The resulting point coordinate image is shown below. Figure 9-2 and 9-2 As shown.

[0210] In the clustering results with k=17, there is a centroid with a very small sample size located in the coordinate range [1.355-1.358, 3.616-3.616], indicating over-segmentation. The clustering results with k=16, however, maintain a more complete energy and spatial correlation, and the centroid distribution matches the data density better, avoiding pseudo-clustering caused by over-segmentation. Therefore, k=16 is chosen as the number of centroids for the three-dimensional clustering, thus determining the number of integrated energy stations to be 16 under the three-dimensional clustering condition. The centroid coordinates and energy consumption data are shown in Table 5.

[0211] Table 5. Coordinates of 3D Cluster Points & Energy Consumption

[0212]

[0213] Substituting the obtained centroid coordinates into ArcMap, we can obtain the comprehensive energy station site selection results under three-dimensional clustering, such as... Figure 10 As shown in the red square.

[0214] The land parcels radiated by the integrated energy station are marked in ArcMap and output as images, such as... Figure 11 As shown.

[0215] The following is a plan for the distributed energy system in the Lingang 103 area, including the site selection of energy stations and the layout of the pipeline network. Figure 12 As shown in Table 6, the energy supply range and pipeline costs of each energy station are listed in Table 7, and the pipe diameters and costs used in the pipeline layout of each energy station are listed in Table 7.

[0216] Table 6 Energy Station Supply Area and Pipeline Costs

[0217]

[0218] Table 7 Pipe diameters and their costs in pipeline network layout.

[0219]

[0220] Based on the aforementioned energy station site selection and power supply range division, hybrid shaping linear programming is used to configure the equipment capacity of the energy stations. The load demand of each energy station is the sum of the loads of energy-consuming buildings within its power supply range, and 8760h load data is used to configure the equipment capacity of the energy stations. Due to the current high price of hydrogen and the cost of proton exchange membrane fuel cells (PEMFCs), there may be situations where the PEMFC configuration is 0. However, with the multiple effects of technological iteration, policy drive, and market expansion in the global hydrogen energy industry, the price of hydrogen and the cost of PEMFCs are rapidly declining. Table 8 shows the changes in PEMFC capacity configuration with fuel cell cost and hydrogen price.

[0221] Table 8. Proton Exchange Membrane Fuel Cell Capacity Configuration

[0222]

[0223] With 2035 as the target scenario, natural gas will remain a stable pillar of the energy system, supporting peak power generation, city gas supply, and traditional industries thanks to mature global infrastructure and industrial inertia. Hydrogen, on the other hand, will become the core engine of the zero-carbon transition. Driven by the reduction in the cost of green hydrogen (lower prices from renewable energy sources + cost reductions in electrolyzers) and the pressure from carbon tax policies, it will aggressively penetrate the fields of fuel cell transportation, hydrogen hybrid power generation, and synthetic fuels. Therefore, it is predicted that by 2035, the ratio of natural gas to hydrogen application will reach 7:3, the cost of proton exchange membrane fuel cells will be 3000 yuan / KW, and the price of hydrogen will be 1.5 yuan / m³. 3 In order to encourage companies to reduce greenhouse gas emissions by increasing carbon emission costs in order to address climate change or promote a low-carbon transition, carbon taxes are gradually increasing. In this scenario, the carbon tax is set at 0.1 yuan / ton, and the costs of each piece of equipment are shown in Table 9.

[0224] Table 9. Equipment Parameters in 2035

[0225]

[0226] Based on the above cost forecasts, the capacity configuration results for various equipment in this region in 2035 are shown in Table 10.

[0227] Table 10. Optimization Results of Energy Station Equipment Capacity Configuration by 2020-2035

[0228]

[0229] As shown in Table 10, the total cost of energy station 8 is relatively high because its peak load is relatively high, which ultimately leads to higher construction and operating costs. On the other hand, the total cost of energy station 31 is relatively low because its peak load is relatively low, which ultimately leads to lower construction and operating costs.

[0230] Further, the annual hydrogen demand of the construction industry in the Lingang 103 area in 2035 was calculated, such as... Figure 13 As shown.

[0231] Depend on Figure 13 It can be seen that the demand for hydrogen is highest in July and lowest in April. This is because July is the hottest part of summer, when the demand for cooling is greater, and electric cooling also leads to increased electricity demand, which in turn increases the output of fuel cells and the demand for hydrogen. In contrast, January is the coldest part of winter, when the demand for heat is greater, which in turn increases the output of fuel cells and auxiliary boilers and the demand for hydrogen. Conversely, the demand for both heat and cold is lower in April and October, and the demand for hydrogen is also lower. Therefore, the curve shows two obvious peaks (summer and winter) throughout the year, with two troughs (spring and autumn) between them, exhibiting a "W"-shaped fluctuation.

[0232] One day each from summer, transitional season, and winter is taken as a typical day. The hydrogen demand for each typical day is shown in the table below. Figure 14 .

[0233] Depend on Figure 14 It can be seen that the hydrogen demand on typical summer and winter days is higher than that on typical days in the transition season. This is because summer and winter have greater demands for cooling and heating, respectively, and the demand from 8:00 to 21:00 is higher than at other times. This is because in summer, this period is the peak of daytime work, commercial activities and residential life, and air conditioning, ventilation and other cooling equipment are running intensively and require a lot of energy to drive them. In winter, this period is the peak of daytime heating (such as heating for office buildings and residential buildings), and heating equipment is running frequently.

[0234] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as defined by the appended claims and their equivalents.

Claims

1. A multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering, characterized in that: The integrated energy system station network multi-layer collaborative optimization method based on GIS three-dimensional clustering includes the following steps: Step 1: Use K-means clustering to select the site for the integrated energy station based on the geographical location and building energy consumption data processed by AcrGIS. Combine the annual time-of-use data of cooling, heating and power load of the integrated energy station to perform equipment configuration and cost calculation based on the three-dimensional clustering site selection results. The second step is to abstract the area under study and establish a pipeline layout planning cost model; solve for the optimal pipeline layout that maximizes the economic efficiency of the pipeline and satisfies the corresponding constraints, and establish a pipeline layout planning cost model. Step 3: Solve the problem based on the pipeline layout plan; Step 4: Conduct coordinated optimization of energy station site selection, pipeline layout, and equipment capacity determination; establish a regional distributed energy system station-network coordinated planning target model; and carry out regional distributed energy system station-network coordinated planning. Step 5: Solve the problem based on the regional distributed energy system station-network collaborative planning.

2. The multi-layer collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 1, characterized in that: The specific steps of the first step include: using ArcGIS to classify building usage types and obtain the building's plot area, center location coordinates, and building energy consumption data; performing K-means clustering analysis on the target data, compiling the optimal k-value candidate range, and using the elbow method to determine the optimal k-value within the range that meets the clustering requirements; obtaining the final clustering results after screening, and substituting the centroid coordinates into ArcMap to obtain the site selection of the integrated energy station under the three-dimensional clustering; performing time-of-use prediction of regional cooling, heating, and electricity loads based on the multi-factor method, and calculating the energy consumption intensity and spatial distribution pattern of each land use type by combining historical data and building type characteristics, thereby obtaining the time-of-use data of the integrated energy station's annual cooling, heating, and electricity loads; and building an equipment configuration and economic analysis model to perform equipment configuration and cost calculations based on the three-dimensional clustering site selection results.

3. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 1, characterized in that: The specific steps of the second step include: the objective function for optimizing the pipeline layout includes four parts, namely the construction cost of the pipeline system, the annual operating cost of the circulating water pump, the energy loss cost of the pipeline system, and the depreciation and maintenance cost of the pipeline system.

4. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 3, characterized in that: The second step is as follows: The construction cost of a pipeline system is related to the pipe diameter and length. For the constructed undirected graph G(V,E,W), this cost can be expressed by the following formula: The pipelines in this section are constructed using direct burial. There are 13 different pipe diameters: DN200, DN250, DN300, DN350, DN400, DN450, DN500, DN600, DN700, DN800, DN900, DN1000, and DN1200. The operating cost of a circulating water pump is related to the pump's design flow rate, electromechanical efficiency, operating pressure, and electricity price, and is calculated using the following formula: Energy loss costs in a pipeline system are directly related to the pipe diameter, length, and insulation thickness. The energy loss of a pipeline system is expressed by the following formula: Depreciation and maintenance costs of pipeline systems: In the actual operation of pipeline systems, there are various foreseeable and unforeseeable losses.

5. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 4, characterized in that: When calculating the economics of a pipeline network, the cost of this component must be considered. This cost is highly uncertain, and in practice, engineers often calculate the cost using a fixed ratio, calculated from the equipment depreciation rate and the maintenance cost ratio, specifically expressed as: In the design of regional distributed energy system pipeline networks, the selection of pipe diameter is influenced by multiple factors, including flow velocity and flow rate. Both flow velocity and flow rate have certain limitations, as detailed below: 0.5<v ij <3.5 (7) 200≤d ij ≤1200 (8)。 6. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 1, characterized in that: The specific steps of the third step include: (301) Construct an undirected graph G(V,E,W) based on the geographic information of the planning area; (302) Input road network information, simulation settings parameters, and energy station location information; (303) Based on the interrelationship of each vertex in network G, solve for the edge matrix E. The specific calculation method is as follows; (304) Input the genetic algorithm control parameters and encode the parameters to optimize the calculation order of energy-consuming buildings; (305) Randomly generate an initial population, set Gen=1, and obtain the Gen generation parent population sample under this layout; (306) Calculate the fitness of the parent population in the Gen generation; (307) Based on this fitness, selection, crossover and mutation are performed to generate a new generation of population, Gen = Gen+1; (308) Repeat steps (305)-(307) until convergence is achieved; (309) Save the optimized calculation order of energy-consuming buildings; (310) Initialize the full adjacency matrix W of the pipeline cost, which is calculated from the previously input parameters as follows: (311) The path of the i-th energy-consuming building is calculated using Dijkstra's algorithm. The path from the energy-consuming building to all energy stations and the increased cost are calculated. The energy station with the lowest cost is selected according to the greedy strategy. Let i = i + 1. (312) Update the cost weight adjacency matrix of the pipeline network. The weight is the increase in pipeline network cost caused by the new access load point. Consider the corresponding economically optimal pipe diameter according to the flow of different pipe segments. The calculation formula is as follows. (313) Repeat steps (11)-(12) until all energy-consuming buildings have been calculated and save the layout and pipe diameter selection of the entire pipeline system; (314) Calculate the objective function C pl ; (315) Calculate the basics, output the objective function value, pipeline layout and pipe diameter of each segment.

7. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 1, characterized in that: The specific steps of the fourth step include: The objective function of the mathematical model for the coordinated planning of regional distributed energy system stations and networks consists of two parts, namely the annualized cost of the energy station and the annualized cost of the pipeline system.

8. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 7, characterized in that: The formulas for calculating the objective function and the energy station cost are as follows: my C total =min(C es,2 +C pl ) (9) The investment cost of an energy station is usually calculated based on the installed capacity of the equipment. The land cost and civil engineering cost of the energy station are usually estimated based on the power load of the energy station.

9. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 8, characterized in that: The investment costs of various energy sources can be expressed by the following equation: Therefore, the total annualized cost of the energy station can be calculated using the following formula:

10. The multi-level collaborative optimization method for integrated energy system station network based on GIS three-dimensional clustering according to claim 1, characterized in that: The specific steps of the fifth step include: (501) Construct an undirected graph G(V,E,W) based on the geographic information of the planning area; (502) Input the parameter information required for the calculation, including genetic algorithm control parameters, economic parameters, etc.; (503) Solve for the edge matrix E based on the interrelationships between the vertices in network G; (504) Determine the upper limit of the number of energy stations and the current number of energy stations; (505) K-means clustering is used to initially determine the location of the energy station, and the location of the energy station is gradually optimized by combining the relative load distance and the energy supply range, and the optimized location of the energy station is saved. (506) Genetic algorithm is used to encode and optimize the calculation order of energy-consuming buildings; (507) Generate an initial population, set Gen=1, and obtain the Gen generation parent population sample. Calculate the fitness of this population. (508) Based on the fitness of each individual in the Gen generation parent population, perform selection, crossover and mutation operations to generate the next generation population sample. Let Gen = Gen+1, and the Gen generation parent population sample is obtained. (509) Determine whether the genetic algorithm has converged at this time. If it has not converged, execute step (506). If it has converged, execute step (510). (510) The optimized energy-consuming building calculation order is obtained after calculation by genetic algorithm; (511) Initialize the cost weight adjacency matrix W and the diameter matrix in network G. Calculate them according to the previous calculation results. The initial values ​​of the diameter matrix are all 0. (512) The current calculation is for the i-th energy-consuming building. Dijkstra's algorithm is used to calculate the path from the energy-consuming building to all energy stations and the corresponding cost increment. The energy station with the lowest cost increment is selected according to the greedy strategy. (513) Based on the selected energy station, record the corresponding path and the flow rate, velocity and pipe diameter of each pipe segment, and update the cost weight adjacency matrix and pipe diameter matrix. The cost value is the increase in network cost caused by the new energy-consuming building, and the pipe diameter is the most economical pipe diameter under this flow rate. (514) Determine whether all energy-consuming buildings have been calculated. If not, proceed to step (512); otherwise, proceed to step (515). (515) Save the obtained pipeline layout and pipe diameter of each pipe section, and calculate the corresponding annualized cost of the pipeline system. (516) Determine the energy supply range of each energy station according to the pipeline network layout, calculate the cooling, heating and power supply capacity of each energy station, and calculate the annualized cost of the energy station. (517) Determine whether the current number of energy stations has reached the upper limit. If not, proceed to step (504); otherwise, proceed to step (518). (518) Output the optimal number of energy stations, and determine the location and energy supply range of the energy stations. Calculate the capacity of the energy station equipment based on the cooling, heating and power load requirements of the energy stations. (519) The calculation ends, and the objective function value, number of energy stations, location of energy stations, pipeline layout, pipeline diameter and energy station equipment capacity are output. Based on the above steps, the solution to the regional distributed energy system station-network coordinated optimization problem is finally obtained.

Images