A hierarchical clustering method based site selection and planning method for distributed district heating system energy stations

Abstract

The present invention discloses a method for site selection and planning of energy stations for distributed regional heating systems based on a hierarchical clustering method. The method comprises the following steps: determining possible locations of distributed energy stations based on the geographical locations of buildings and streets within the heating area; determining typical heat loads of buildings within the planning area based on the geographical location distribution of buildings requiring heating within the heating area and the heat loads required by the buildings; using a hierarchical clustering algorithm to determine possible location scheme combinations for distributed energy stations; solving constraints based on thermal and hydraulic constraints for possible location scheme combinations for distributed energy stations, and using a double-layer optimization solution to obtain the optimal results of possible schemes with the goal of minimizing the initial investment cost of the pipeline network; and completing the site selection of distributed energy stations based on the optimal results of all possible schemes, taking the scheme with the minimum initial investment cost within the area as the final scheme. The present invention can be widely applied to the technical field of optimization and planning of heating systems.

Description

Technical Field

[0001] The present invention relates to the field of heating system planning and design, and in particular to a distributed regional heating system energy station site selection and planning method based on a hierarchical clustering method. Background Art

[0002] While significant progress has been made in the planning and design of heating systems, planning and design methods for emerging scenarios and demands have yet to fully catch up. Specifically, most heating system planning and design methods simply optimize the network topology based on the location and number of heat sources, or directly use the minimum spanning tree algorithm to generate the shortest path in the system without considering the actual construction conditions of the network.

[0003] To meet the demand for local consumption of renewable energy in the heating sector, the current trend in heating systems is towards miniaturization. The shift from a centralized to a decentralized heating system raises a series of technical challenges during the planning and design process. Smaller energy stations facilitate regional regulation, simplify pipeline network construction, and improve control over thermal and hydraulic conditions. However, economies of scale inevitably lead to increased initial investment costs per unit area. Therefore, during heating system construction, it is crucial to fully consider the optimal system size. Therefore, planning a heating system requires careful consideration not only of the number and location of energy stations but also of optimizing the pipeline network based on actual conditions. Summary of the Invention

[0004] In view of the defects existing in the above-mentioned background technology, the purpose of the present invention is to provide a distributed district heating system energy station site selection planning method based on a hierarchical clustering method, thereby providing guiding opinions for the site selection planning of the heating system energy station.

[0005] In order to achieve the above object, the technical solution proposed by the present invention is as follows:

[0006] A method for site selection and planning of energy stations in distributed district heating systems based on hierarchical clustering method, including

[0007] Step S1: Determine the possible locations of distributed energy stations based on the geographical locations of buildings and streets in the heating area;

[0008] Step S2: Determine the typical heat load of buildings in the planning area based on the geographical distribution of buildings requiring heating within the heating area and the heat load required by the buildings; and use a hierarchical clustering algorithm to determine possible location combinations for distributed energy stations.

[0009] Step S3: Based on the possible location schemes for distributed energy stations and the thermal-hydraulic constraints, the constraints are solved. With the goal of minimizing the initial investment cost of the pipeline network, a two-level optimization solution is used to obtain the optimal result of the possible schemes. The scheme specifically includes the topological structure and pipe diameter of the designed heat transmission pipelines in the heating area.

[0010] Step S4: Based on the optimal heat transmission pipeline topology and pipe diameter of each scheme designed in the heating area, determine the scheme with the lowest initial investment cost in the area and complete the site selection of the distributed energy station.

[0011] In the above technical solution, further, step S1 is specifically to select the possible location of the distributed energy station according to the actual case topology structure, the energy station location requirements in the building design standards and the actual situation, which should generally be near the building at the boundary of the area.

[0012] Furthermore, in step S2:

[0013] The typical heat load size of buildings in the planning area is determined based on the geographical distribution of buildings that need heating in the heating area and the heat load size required by the buildings. The specific method is as follows: buildings in similar areas are divided into building scenes based on building type, building location information, and building construction age, and then meteorological data scenes are divided according to weather conditions, so as to achieve complete building scene division, and the typical heat load size of the building is determined based on the divided building scenes, and then the typical heat load size of the buildings in the planning area is obtained.

[0014] The hierarchical clustering algorithm is used to determine possible location scheme combinations for distributed energy stations. The specific process is as follows: the geographical location of the building is represented by two-dimensional coordinates, the heat load size of the building is dimensionless, and then the two are used as inputs to calculate possible schemes using the hierarchical clustering algorithm. In the hierarchical clustering algorithm, the Euclidean distance is calculated using the following formula (1), and the distance between different clusters is calculated using the Ward method, as shown in the following formula (2). Based on the distance between different clusters in the clustering algorithm, the appropriate number of clusters and the possible locations of the distributed energy stations are screened to determine the possible location scheme combinations for the distributed energy stations.

[0015] The Euclidean distance d between building i and building j (i,j) The calculation formula is:

[0016]

[0017] Where p is the total dimension of the input building data, which is 3; k represents the kth dimension; i and j represent the i-th building and j-th building of the input, respectively; x ik and x jkInput data of the kth dimension for the i-th and j-th buildings respectively;

[0018] For all building data, the calculation formula of the Ward distance L(r,s) between cluster s and cluster r in the same dimension is:

[0019]

[0020] Where n r represents the number of buildings in cluster r, n s represents the number of buildings in cluster s, x ri and x sj are the input data of building i in cluster r and building j in cluster s under the same dimension; d(x ri ,x sj ) represents x ri and x sj The Euclidean distance between .

[0021] Furthermore, in step S3, the constraints are as follows:

[0022] C n =Q (n,j),invest c invest +y (n,j) c investfix (3)

[0023] Q loss,n,t =Q (n,j),invest ·f loss,t +y (n,j) ·f lossfix,t (4)

[0024] Q (n,j),t =Q (i,n),t -Q loss,n,t (5)

[0025] Q (i,n),invest =Q (n,j),invest (6)

[0026]

[0027] -Q (i,n),invest ≤Q (i,n),t ≤Q (i,n),invest (8)

[0028] -Q (n,j),invest ≤Q (n,j),t ≤Q (n,j),invest (9)

[0029] y n,j ∈{0,1} (10)

[0030] Where Cn is the initial investment in the pipeline network, CNY; Q (n,j),invest is the heat transfer capacity of the pipe network n, kW; Q (n,j),t is the heat flow out of pipe network n at time t, kW; Q (i,n),t is the heat flow into the pipe network n at time t, kW; Q loss,n,t is the heat loss of pipe network n at time t, kW; y(n,j) is the decision variable, 0 represents no pipe network, 1 represents the presence of a pipe network; c invest is the initial investment cost per unit input, CNY / kW; c investfix is fixed investment expenditure, CNY; f loss,t is the cost coefficient of heat transfer capacity at time t in kW / kW; f lossfix,t is the cost coefficient of fixed investment expenditure at time t, kW / kW; is the minimum pipe network transport limit, kW, Maximum pipeline network transmission limit, kW;

[0031] The objective function is to minimize the initial investment cost of the pipeline network, and the calculation formula is:

[0032] Obj=min∑C n (11)

[0033] A two-level optimization solution is used to obtain the optimal result of possible solutions. Specifically, the upper level uses mixed integer linear programming to solve the topological structure of the heat transmission pipeline designed in the heating area, and the lower level uses linear programming to solve the diameter of each pipe section.

[0034] Furthermore, the cbc solver is used to solve the objective function.

[0035] Furthermore, in step S4, according to the initial investment of regional energy stations of different sizes given in the "Municipal Engineering Investment Estimation Indicators", the initial investment corresponding to the unit area is calculated, and finally the cost curves of distributed energy stations of different sizes are fitted. Then, the size of the distributed energy station corresponding to the solution solved in step S3 is substituted into the cost curve to calculate the initial investment of each distributed energy station. Finally, according to formula (12), the solution with the minimum initial investment cost in the final region is calculated to complete the site selection of the distributed energy station.

[0036] C=C p +C h (12)

[0037] Where C p and C h They are the investment costs of the pipeline network and the investment costs of the distributed energy station respectively.

[0038] The beneficial effects of the present invention are:

[0039] It is very difficult to directly plan and design the heating system based solely on the system's geometric location information and the size of the heat load required by the heat users during the planning and design phase. This is because there are many feasible combinations of energy station locations and quantities. Therefore, in order to reduce the number of feasible solutions and reduce the time cost of planning and design, the present invention innovatively uses a hierarchical clustering algorithm, and at the same time, combines the actual situation to obtain a feasible location solution for the distributed energy station. Furthermore, based on the feasible location solution obtained above, the present invention uses mixed integer linear programming and linear programming to design the optimal pipe network structure and the pipe diameter size of each section of the pipe network under the corresponding solution. Finally, combined with the energy station cost curve obtained by fitting, the investment and construction cost of each solution is finally obtained. By comprehensively comparing the various solutions, the optimal solution for the energy station of the distributed regional heating system is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings in the specification, which constitute a part of this application, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0041] In the attached figure:

[0042] Figure 1 It is a step diagram for implementing the overall planning method of the present invention.

[0043] Figure 2 Schematic diagram of the Ward distance calculation of the present invention.

[0044] Figure 3 It is a heat flow diagram of the pipe network structure.

[0045] Figure 4 It is a schematic diagram of the hierarchical clustering results.

[0046] Figure 5 This is a schematic diagram of the heating area. DETAILED DESCRIPTION

[0047] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The exemplary embodiments and descriptions are only used to explain the present invention but are not intended to limit the present invention.

[0048] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0049] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0050] like Figure 1 As shown; the completed planning process in the present invention includes the following four steps, and the purpose of each step is: first, determining the possible locations of distributed energy stations; second, determining the possible location scheme combinations of distributed energy stations; third, determining the optimal heat transmission pipeline network topology and pipe diameter size under each scheme; finally, determining the scheme with the lowest initial investment cost.

[0051] First, determine the possible location of the distributed energy station based on the geographical location of buildings and streets in the heating area, including selecting the possible location of the distributed energy station based on the actual case topology, the energy station location requirements in the building design standards and the actual situation. Generally, the distributed energy station should be near the buildings at the boundary of the area.

[0052] Second, based on the geographical distribution of buildings that need heating in the heating area and the required heat load of the buildings, the specific process of determining the typical heat load of buildings in the planning area is as follows: buildings in similar areas are preliminarily divided into building scenes based on building type, building location information, and building construction year, and then meteorological data scenes are divided according to weather conditions to achieve complete building scene division. The typical heat load of the building is determined based on the divided building scenes, and then the typical heat load of the buildings in the planning area is obtained.

[0053] The hierarchical clustering algorithm is used to determine the possible location scheme combinations of distributed energy stations. The specific process is as follows: the geographical location of the building is represented by two-dimensional coordinates, the heat load size of the building is dimensionless, and then the two are used as input to calculate the possible schemes using the hierarchical clustering algorithm; in the hierarchical clustering algorithm: the Euclidean distance between different points is calculated as follows (1), and the distance between different clusters is calculated using the Ward method as follows (2). The schematic diagram is as follows Figure 2 According to the distance between different clusters in the clustering algorithm, the appropriate number of clusters and the possible locations of distributed energy stations are screened out, and the possible location scheme combinations of distributed energy stations are determined.

[0054] The Euclidean distance d between building i and building j (i,j) The calculation formula is:

[0055]

[0056] Where p is the total dimension of the input building data, which is 3 here; k represents the kth dimension; i and j represent the i-th building and j-th building in the input respectively; x ik and x jk Input data of the kth dimension for the i-th and j-th buildings respectively;

[0057] For all building data, the calculation formula of the Ward distance L(r,s) between cluster s and cluster r in the same dimension is:

[0058]

[0059] Where n r represents the number of buildings in cluster r, n s represents the number of buildings in cluster s, x ri and x sj are the input data of building i in cluster r and building j in cluster s under the same dimension; d(x ri ,x sj ) represents x ri and x sj The Euclidean distance between .

[0060] Third, based on the thermal-hydraulic constraints, we solved the possible location combinations for distributed energy stations. With the goal of minimizing the initial investment cost of the pipeline network, we used a two-level optimization solution to obtain the optimal solution. The solution specifically included the topology and diameter of the heat pipes designed within the heating area.

[0061] According to the possible location scheme combinations of distributed energy stations, according to the thermal hydraulic constraints, the constraints are solved, and the initial investment of the scheme pipeline network is minimized. The optimal result of the possible scheme is obtained by using a two-level optimization solution. Specifically, the upper layer uses mixed integer linear programming to solve the topological structure of the heat transmission pipeline designed in the heating area, and the lower layer uses linear programming to solve the diameter size of each pipe section. The optimized pipeline model is shown in the attached figure. Figure 3 The open-source DHNx module is called using the CBC solver to solve the mixed linear integer programming and linear scaling models for each solution.

[0062] The constraints are as follows:

[0063] C n =Q (n,j),invest c invest +y (n,j) c investfix (3)

[0064] Q loss,n,t =Q (n,j),invest ·f loss,t +y(n,j) ·f lossfix,t (4)

[0065] Q (n,j),t =Q (i,n),t -Q loss,n,t (5)

[0066] Q (i,n),invest =Q (n,j),invest (6)

[0067]

[0068] -Q (i,n),invest ≤Q (i,n),t ≤Q (i,n),invest (8)

[0069] -Q (n,j),invest ≤Q (n,j),t ≤Q (n,j),invest (9)

[0070] y n,j ∈{0,1} (10)

[0071] Where C n is the initial investment in pipeline network n [CNY], Q (n,j),invest is the heat transfer capacity of the pipe network n [kW], Q (n,j),t is the heat flow [kW] outflowing from pipe network n at time t, Q (i,n),t is the heat flow [kW] flowing into the pipe network n at time t, Q loss,n,t is the heat loss of pipe network n at time t [kW], y(n,j) is the decision variable (0 means no pipe network, 1 means pipe network). invest is the initial investment cost per unit input [CNY / kW], c investfix is fixed investment expenditure [CNY], f loss,t is the cost coefficient [kW / kW] that depends on the heat transfer capacity at time t, f lossfix,t is the cost coefficient [kW / kW] of fixed investment expenditure at time t, is the minimum network transport limit [kW], Maximum pipeline network transmission limit [kW].

[0072] The objective function is to minimize the initial investment cost of the pipeline network, and the calculation formula is:

[0073] Obj=min∑C n (11)

[0074] Finally, based on the optimal results of all possible solutions obtained in step S3, the solution with the lowest initial investment cost in the region is selected as the final solution to complete the site selection of the distributed energy station. Specifically:

[0075] According to the initial investment of regional energy stations of different sizes given in the "Municipal Engineering Investment Estimation Indicators", the initial investment corresponding to the unit area is calculated, and finally the cost curves of distributed energy stations of different sizes are fitted. Then, the size of the distributed energy station corresponding to the solution obtained in step S3 is substituted into the cost curve to calculate the initial investment of each distributed energy station. Finally, according to formula (12), the solution with the minimum initial investment cost in the final region is calculated to complete the site selection of the distributed energy station.

[0076] C=C p +C h (12)

[0077] Where C p and C h They are investment in pipeline networks and investment in distributed energy stations.

Claims

1. A method for site selection and planning of energy stations in a distributed district heating system based on a hierarchical clustering method, characterized in that: The following steps are involved: Step S1: Determine the possible locations of distributed energy stations based on the geographical locations of buildings and streets in the heating area; Step S2: Based on the geographical distribution of buildings requiring heating within the heating area and the required heat load of the buildings, the typical heat load of the buildings within the planning area is determined; a hierarchical clustering algorithm is used to determine possible location scheme combinations for distributed energy stations. The specific method is as follows: buildings in similar areas are preliminarily divided into building scenarios based on building type, building location information, and building construction year. Meteorological data scenarios are then divided based on weather conditions to achieve complete building scenario division. The typical heat load of the buildings is determined based on the divided building scenarios, thereby obtaining the typical heat load of the buildings within the planning area. The geographical location of the building is represented by two-dimensional coordinates, and the heat load of the building is dimensionless. Then, the two are used as inputs to calculate possible solutions using a hierarchical clustering algorithm. In the hierarchical clustering algorithm, the Euclidean distance between different points is calculated using the following formula (1), and the distance between different clusters is calculated using the Ward method using the following formula (2). According to the distance between different clusters in the clustering algorithm, the appropriate number of clusters and the possible locations of distributed energy stations are screened out, and the possible location scheme combinations of distributed energy stations are determined; The Euclidean distance d between building i and building j (i,j) The calculation formula is: Where p is the total dimension of the input building data, which is 3; k represents the kth dimension; i and j represent the i-th building and j-th building of the input, respectively; x ik and x jk Input data of the kth dimension for the i-th and j-th buildings respectively; For all building data, the calculation formula of the Ward distance L(r,s) between cluster s and cluster r in the same dimension is: Where n r represents the number of buildings in cluster r, n s represents the number of buildings in cluster s, x ri and x sj are the input data of building i in cluster r and building j in cluster s under the same dimension; d(x ri ,x sj ) represents x ri and x sj The Euclidean distance between Step S3: For possible location combinations of distributed energy stations, solve the constraints based on thermal hydraulic constraints. With the goal of minimizing the initial investment cost of the pipeline network, use a two-level optimization solution to obtain the optimal result of the possible solutions. The solution specifically includes the designed heat transmission pipeline topology and pipe diameter within the heating area. The specific constraints are as follows: C n =Q (n,j),invest ·c invest +y (n,j) ·c investfix (3) Q loss,n,t =Q (n,j),invest ·f loss,t +y (n,j) ·f lossfix,t (4) Q (n,j),t =Q (i,n),t -Q loss,n,t (5) Q (i,n),invest =Q (n,j),invest (6) -Q (i,n),invest ≤Q (i,n),t ≤Q (i,n),invest (8) -Q (n,j),invest ≤Q (n,j),t ≤Q (n,j),invest (9) y n,j ∈{0,1} (10) Where C n is the initial investment in the pipeline network, CNY; Q (n,j),invest is the heat transfer capacity of the pipe network n, kW; Q (n,j),t is the heat flow out of pipe network n at time t, kW; Q (i,n),t is the heat flow into the pipe network n at time t, kW; Q loss,n,t is the heat loss of pipe network n at time t, kW; y(n,j) is the decision variable, 0 represents no pipe network, 1 represents the presence of a pipe network; c invest is the initial investment cost per unit input, CNY / kW; c investfix is fixed investment expenditure, CNY; f loss,t is the cost coefficient of heat transfer capacity at time t in kW / kW; f lossfix,t is the cost coefficient of fixed investment expenditure at time t, kW / kW; is the minimum pipe network transport limit, kW, Maximum pipeline network transmission limit, kW; The objective function is to minimize the initial investment cost of the pipeline network, and the calculation formula is: Obj=min∑C n (11) A two-level optimization solution was used to obtain the optimal solution. Specifically, the upper level used mixed integer linear programming to solve the topological structure of the heat pipes designed within the heating area, while the lower level used linear programming to solve the diameter of each pipe segment. Step S4: For the optimal results of all possible solutions obtained in step S3, the solution with the minimum initial investment cost in the region is taken as the final solution to complete the site selection of the distributed energy station; according to the initial investment of regional energy stations of different sizes given in the "Municipal Engineering Investment Estimation Indicators", the initial investment corresponding to the unit area is calculated, and finally the cost curves of distributed energy stations of different sizes are fitted, and then the scale of the distributed energy station corresponding to the solution solved in step S3 is substituted into the cost curve to calculate the initial investment of each distributed energy station, and finally the solution with the minimum initial investment cost in the region is calculated according to formula (12), and the site selection of the distributed energy station is completed; C=C p +C h (12) Where C p and C h They are the investment costs of the pipeline network and the investment costs of the distributed energy station respectively.

2. A method for site selection and planning of energy stations in a distributed district heating system based on a hierarchical clustering method according to claim 1, characterized in that: Specifically, step S1 is to select possible locations of distributed energy stations according to the actual case topology, energy station location requirements in the building design standards, and actual conditions.

3. A method for site selection and planning of energy stations in a distributed district heating system based on a hierarchical clustering method according to claim 1, characterized in that: Use the cbc solver to solve the objective function.

Images