Water-energy-carbon coupling based water pipe network planning method and system based on double-layer planning and graph algorithm

Abstract

The present application relates to a kind of water-pipe network planning method and system based on double-layer planning and graph algorithm under water-energy-carbon coupling, comprising: step 1: measurement and prediction data: make area simplified graph;Predict user daily water demand, and triangular fuzzy membership function;Estimate the power consumption of each candidate pipeline;Measure the pipe length needed when laying pipe in each candidate road;Step 2: model is established, model includes the upper layer objective function of minimizing annual total cost and the lower layer objective function of minimizing daily power consumption of pipe network;Step 3: based on double-layer planning and graph algorithm, the optimal value of the model established in step 2 is obtained, and the optimal water pipe network planning scheme is obtained.The present application can effectively quantify the energy consumption in water pipe network system, balance the contradiction between economic saving and environmental protection, while maintaining the main position of economic demand, provide scientific and reasonable decision-making scheme for decision-makers with different needs in water plant construction, pipeline layout, pipeline load.

Description

Technical Field

[0001] This invention relates to a water network planning method and system based on bi-level programming and graph theory algorithms under the coupling effect of water, energy and carbon, belonging to the field of water resources planning technology. Background Technology

[0002] Since the Second Industrial Revolution, global energy resources have been widely used in various fields. However, the reserves of non-renewable primary energy sources such as coal and oil are limited. In recent years, with the large-scale exploitation and irrational use of resources, the problem of energy shortages has gradually become apparent. Simultaneously, people's water demand in industry and daily life has been gradually increasing. The unscientific discharge of wastewater has led to the pollution of rivers, lakes, and other water bodies, exacerbating water scarcity. It is worth noting that water, energy, and carbon emissions are not unrelated. The 2012 World Water Forum proposed a hydropower framework, and the 2014 World Water Day presented a report on water and energy, gradually attracting the attention of the scientific community to the coupling relationship between water and energy. Subsequently, with the signing of the Paris Agreement, the control of carbon emissions has gradually gained importance among countries. It is noteworthy that water consumes energy and generates carbon emissions during extraction, transportation, consumption, discharge, and treatment. It is estimated that approximately 15% of global water production is used in the energy industry, and approximately 8% of global energy is used in water production, consumption, and treatment. Furthermore, greenhouse gas emissions cannot be ignored throughout the energy consumption lifecycle; studies often quantify related greenhouse gas emissions using carbon dioxide equivalents. As a result, the study of the water-energy-carbon coupling relationship has gradually gained attention in various fields.

[0003] During the construction and operation of water pipeline networks, a significant amount of energy is consumed in building material production, water extraction and purification, and water distribution, resulting in carbon emissions. However, most existing water pipeline network planning problems use economic objectives as the sole metric for optimal solutions, or calculate energy consumption based on economic costs, such as the Improved Bat algorithm proposed by Lyu et al. and the A Graph Theory-Based Method proposed by Chen and Zhu. These methods do not comprehensively consider energy and carbon emissions, weaken the contradiction between economics, energy, and the environment, and are not conducive to decision-makers' overall understanding of the system's environmental impact.

[0004] With the progress of civilization, environmental issues have received increasing attention, and scholars in the field of water network planning have gradually emphasized the study of the environmental impact of systems. For example, Herstein et al. developed a multi-objective programming model for water distribution systems, with energy consumption as one of the objectives. However, this model did not consider prioritizing economic development over environmental factors, nor did it optimize the spatial layout of pipelines. In recent years, with the government's goals of carbon peaking and carbon neutrality, water network planning should pay more attention to the adverse environmental impacts of energy consumption and carbon emissions, and choose planning schemes that reduce energy consumption and carbon emissions. There is relatively little analysis and research on the coupling effects of water, energy, and carbon in water network planning. Although Wu et al. proposed an optimization scheme that applies a multi-objective genetic algorithm to the influent distribution network, considering the impact of carbon emissions on the system, they neglected the significant impact of energy on the system.

[0005] In summary, existing water pipeline planning methods suffer from several problems: their optimization objectives are relatively singular, their consideration of environmental impact is vague, their analysis of the contradiction between energy consumption and economic conservation is insufficient, they cannot account for hidden energy consumption during pipeline construction (such as energy consumption during pipeline production), and they fail to consider current societal constraints on carbon emissions. Therefore, they are not conducive to decision-making adjustments by decision-makers with different needs. While environmental protection (such as low-carbon and emission reduction) is considered as an objective in multi-objective water pipeline models, it overlooks the fact that current economic development remains the most important goal, and environmental protection is another crucial factor based on economic development. These models often provide numerous candidate solutions for decision-makers to decide freely, leaving them without a basis for final selection. Furthermore, existing planning methods often use quantifiable variables such as load as optimization objects, with limited research on spatial layout optimization methods. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a water network planning method based on bi-level programming and graph theory algorithms under the coupling effect of water, energy, and carbon. This invention is dedicated to the optimized design of water network layout, water plant construction, and water supply under carbon emission reduction requirements.

[0007] This invention also provides a water pipeline planning system based on bi-level programming and graph theory algorithms under water-energy-carbon coupling.

[0008] Terminology Explanation:

[0009] 1. Floyd's algorithm is a computer algorithm used to calculate the shortest path between two points in a graph. Graph G consists of point v i ∈V and edge e i ∈E constitutes, if there exists a vertex i(v i ) to vertex j(v jThe method involves connecting vertex i and vertex j with an edge. Two vertices are said to be adjacent if they are connected by at least one edge. Let W be the weight matrix of the graph (weights, i.e., values, represent costs, distances, etc. in the practical problem), w ij Record the weights from vertex i to vertex j. A path is formed by edges between adjacent vertices. The shortest path d between vertex i and vertex j is... ij The Floyd algorithm finds the path with the smallest sum of weights among all paths, starting from vertex i and ending at vertex j. The basic calculation steps are as follows:

[0010] Step 1: Input the weight matrix W of the graph. For all i and j, we have d ij =w ij , k = 1.

[0011] Step 2: Update d ij For all i and j, if d ik +d kj <d ij Let d ij =d ik +d kj .

[0012] Step 3: Stop when k = n, otherwise go to step 2.

[0013] 2. The Improved Prim algorithm, proposed by Chen and Zhu in their paper "A Graph Theory-Based Method for Regional Integrated Energy Network Planning: A Case Study of a China–USLow-Carbon Demonstration City," is an improved minimum spanning tree algorithm based on the Prim algorithm. A graph connected to any two points without cycles is called a tree. A tree consisting of some edges and all vertices in the graph is called a spanning tree. The spanning tree with the minimum sum of edge weights is the minimum spanning tree. In waterway network load calculation, the pipeline load is determined by the total load of downstream users; therefore, the edge weights are affected by the network layout. The traditional Prim algorithm cannot calculate the minimum spanning tree of a waterway network, hence the improvement. The steps of the Improved Prim algorithm are as follows:

[0014] Step 1: Input the starting point v0, the weight matrix W, and the set of points U∈V that must be included in the spanning tree. Assign the set of points recording the minimum spanning tree to v0 (T=v0). Assign the set of points recording the connection order of the minimum spanning tree to the zero set (P=zeros). Denote the distance from the starting point to all other points as infinity, and the distance from the starting point to itself as 0 (TL={tl(v0)=0, ... i )=inf|v i ∈V\v0}).

[0015] Step 2: Record all edges where one vertex belongs to T and another vertex does not belong to T as L = {l ij |v i ∈T&v j ∈(V\T)}. Calculate the current shortest distance for all points.

[0016] Step 3: Place l ij All endpoints v j Current distance Ttl(v) j The smallest point v k =min{tl(v j )|l ij ∈L} is recorded in set T. Record the connection v. k Point P(v) k )={v i |cm(v k )=tl(v i )+w ik}

[0017] Step 4: Assign value to tl(v) k )=cm(v k Clear set L.

[0018] Step 5: If all elements in set U belong to set T, then end the algorithm; otherwise, return to step 2.

[0019] The technical solution of this invention is as follows:

[0020] A water pipeline network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling effect includes the following steps:

[0021] Step 1: Measure and predict data, specifically including:

[0022] Create a simplified map of the area;

[0023] The daily water demand of users is predicted based on their historical water consumption, and represented by a triangular fuzzy membership function.

[0024] Based on the height of the control point above the water surface of the pump station's suction well, the minimum service head required at the control point, and the total head loss from the control point to the pump station, estimate the power consumption of each candidate pipeline segment.

[0025] Measure the required pipe length for laying pipes on each candidate road segment;

[0026] Step 2: Establish a model, which includes an upper-level objective function that minimizes the total annual cost and a lower-level objective function that minimizes the daily electricity consumption of the pipeline network;

[0027] Step 3: Based on bi-level programming and graph theory algorithms, find the optimal value of the model established in Step 2 and obtain the optimal water pipe network planning scheme.

[0028] According to a preferred embodiment of the present invention, in step 1, creating a simplified regional map specifically refers to:

[0029] Candidate water plant locations, user locations, and transportation hub locations are simplified to solid squares, solid circles, and dots, respectively. Users are categorized into six types: energy stations, commercial office buildings, residential buildings, hotels, educational buildings, and government buildings, with simple distinctions made. Candidate roads for pipeline laying are simplified to lines.

[0030] According to a preferred embodiment of the present invention, in step 1, the membership function μ q The mathematical expression for (x) is shown in equation (Ⅰ):

[0031]

[0032] In equation (Ⅰ), q p The minimum possible water demand value for the user, expressed in meters (m). 3 ;q o The maximum possible water demand value for the user, in meters. 3 ;q m The water demand value for the user with the highest probability, in meters. 3 ;q p and q o The corresponding membership degree is 0, q m The corresponding membership degree is 1.

[0033] According to a preferred embodiment of the present invention, in step 1, the power consumption of each candidate pipeline segment is estimated based on the height Z (m) of the ground level of the control point above the water surface of the pump station's suction well, the minimum service head F (m) required for the control point, and the total head loss h (m) from the control point to the pump station; the specific estimation method is as follows:

[0034] Pipeline e i Electricity consumption per cubic meter of water transport ρ is the density of water, which is 1000 kg / m³. 3g is the acceleration due to gravity, which is 9.81 m / s². 2 η represents the overall efficiency of the pumping station; Z i pipe segment e i The height difference between the start and end points, F i pipe segment e i Minimum service head, h i pipe segment e i Head loss.

[0035] According to a preferred embodiment of the present invention, in step 2, the mathematical expression of the upper-level objective function that minimizes the total annual cost is shown in equation (II):

[0036] TC = WPC + PNC (II)

[0037] In formula (II), TC refers to the annual total cost, WPC refers to the annual cost of the water plant, and PNC refers to the annual cost of the pipeline network.

[0038]

[0039]

[0040] in, WP (representing the water plant) j The minimum possible daily load is calculated by summing the minimum possible user water demand values, i.e. WP (representing the water plant) j The daily load value with the highest probability is calculated by summing the water demand values ​​of the users with the highest probability. WP (representing the water plant) j The maximum possible daily load is calculated by summing the maximum possible user water demand values, i.e. m represents the total number of water plants designed; f(q) represents the construction cost of the water plants, through... Calculations are performed, SMP is the estimated parameter; T a This indicates the payback period for the water plant construction investment, in years; PRW j Indicates water plant WP j Production cost per cubic meter of water; IPC represents the investment cost of pipeline construction, in yuan; T2 represents the payback period for pipeline construction investment; Y1 represents the annual depreciation and overhaul cost of the pipeline, in yuan / year, calculated according to a fixed ratio of the pipeline construction investment cost; Y2 represents the annual operating cost, in yuan / year, mainly calculating the operating electricity cost.

[0041] According to a preferred embodiment of the present invention, in step 2, the mathematical expression of the lower-level objective function for minimizing the daily power consumption of the pipeline network is shown in equation (Ⅲ):

[0042] TE = ECN + EWP + EWD (III)

[0043] In formula (Ⅲ), TE refers to the daily electricity consumption of the water pipeline network; ECN refers to the electricity consumption for pipeline production; EWP refers to the electricity consumption for water production; and EWD refers to the electricity consumption for water transmission in the pipeline network.

[0044] The calculations for ECN, EWP, and EWD are as follows:

[0045]

[0046]

[0047]

[0048] Where EFC is the pipeline energy consumption coefficient, the unit is kWh / kg; PG i For pipe weight; EQW j The electricity consumption per cubic meter of water produced by the water plant;

[0049] According to a preferred embodiment of the present invention, in step 2, the constraints of the upper-level objective function for minimizing the total annual cost and the lower-level objective function for minimizing the daily power consumption of the pipeline network include the following:

[0050] a. Each user is supplied with water by only one water plant;

[0051] b. Carbon emission limits are calculated as follows: CEF is the carbon emission factor for electricity production;

[0052] The confidence level for carbon emission constraints on water supply is at least 50%, i.e., α ≥ 0.5;

[0053] c. Satisfaction level cannot be less than 0 or greater than 1; upper-level decision-makers have the decision-making power to set the lower limit of satisfaction level (MS).

[0054] The calculation method for the overall satisfaction index λ between decision-makers at both upper and lower levels is as follows:

[0055] λ=min{A,B}; MS≤A≤1; 0≤B≤1;

[0056]

[0057]

[0058] in, Representing the minimum economic cost when only an economic objective is achieved, according to The energy target value f′ corresponding to the minimum economic cost is calculated based on the optimal solution. L Then f′ LThe highest energy consumption acceptable to lower-level decision-makers; This refers to the minimum energy consumption required for the energy target, and is based on... The optimal solution is used to determine the highest acceptable economic cost f′ for the upper-level decision-maker. U .

[0059] Further preferred, MS = 0.3.

[0060] According to a preferred embodiment of the present invention, in step 3, the optimal value of the model established in step 2 is obtained based on a bi-level programming and graph theory algorithm to obtain the optimal water pipeline network planning scheme. The specific implementation steps include:

[0061] (1) Calculate the optimal solution and optimal value when only considering the upper-level objective function. The optimal solution considering only the upper-level objective function is the pipeline layout, user zoning, and water plant construction scheme adopted to minimize the total annual cost; optimal value. That is, the minimum total annual cost; and the corresponding lower-level objective value f′ is obtained using the optimal solution. L The lower-level target value f′ L That is, the daily power consumption of the pipeline network;

[0062] Calculate the optimal solution and optimal value when only the lower-level function is considered. The optimal solution considering only the lower-level function refers to the pipeline layout, user zoning, and water plant construction scheme adopted to minimize daily power consumption; optimal value. That is, the minimum daily power consumption of the pipeline network, and the corresponding upper-level target value f′ is obtained using the optimal solution. U Target value f′ U That is, the total annual cost;

[0063] (2) The adjacency matrix and membership function of the simplified input region graph. p q m q o Three values, the candidate pipe segment water transport power consumption estimation matrix EC, and the candidate pipe segment length matrix;

[0064] An adjacency matrix is ​​a matrix that indicates whether a pipeline can be laid.

[0065] The candidate pipe segment length matrix LH refers to the matrix composed of the straight-line distances from the start point to the end point of the pipe segment;

[0066] (3) Enumerate all possible combinations of water plants: by SK K N The set records all possible combinations including K candidate water plants, N = 1, 2, ..., K. N ;

[0067] For each combination, perform a search. Once all combinations have been searched, the search ends. Otherwise, proceed to step (4) to continue the search. The search process is as follows:

[0068] (4) The shortest water transmission distance between the user and all water plants is calculated by the Floyd algorithm and recorded as matrix SW;

[0069] (5) Assign all users to the water plants in the search combination according to matrix SW, and record the assignment results into matrix SPW;

[0070] (6) Calculate the minimum total load pipeline layout scheme W using the Improved Prim algorithm. r According to the different zones of the water supply plants, the water supply plants, the users supplied by the plants, and the laying path of the water transmission pipeline during the water supply process are recorded in zone P. r middle;

[0071] (7) Calculate the carbon emission constraint confidence level α. If α≤0.5, mark the scheme as infeasible, flag=1, and set N to increase by 1, λ=0, flag=0, and return to step (4) to retrieve the next water plant combination scheme; if α>0.5, proceed to step (8);

[0072] (8) Calculate the target values ​​f for the upper and lower layers corresponding to the calculation scheme. U =TC,f L =TE; Overall satisfaction of decision-makers at both upper and lower levels λ = min{A,B};

[0073] (9) If the overall satisfaction is negative, the scheme is marked as infeasible, flag=1, and N is increased by 1, λ=0, flag=0, and return to step (4) to retrieve the next water plant combination scheme; if λ is positive, proceed to step (10);

[0074] (10) Determine whether the current solution is the optimal solution, i.e., the solution corresponding to the maximum value of the overall satisfaction λ. If so, record the current solution f. U f L , λ and P r Otherwise, it will not be recorded;

[0075] (11) Set N to increase by 1, λ = 0, flag = 0, and return to step (4) to retrieve the next water plant combination scheme.

[0076] A water pipeline network planning system based on bi-level programming and graph theory algorithm under water-energy-carbon coupling effect, including a data measurement and prediction unit, a model building unit and an optimal water pipeline network planning scheme acquisition unit;

[0077] The data measurement and prediction unit is used to perform step 1; the model building unit is used to perform step 2; and the optimal water network planning scheme determination unit is used to perform step 3.

[0078] The beneficial effects of this invention are as follows:

[0079] 1. This method can effectively quantify energy consumption in water pipeline systems, balance the contradiction between economic conservation and environmental protection, and maintain the primary position of economic demand, providing scientific and reasonable decision-making solutions for decision-makers with different needs in terms of water plant construction, pipeline layout, and pipeline load.

[0080] 2. Compared to single-objective water network planning models, the model algorithm of this invention, while achieving spatial optimization, more accurately quantifies the energy consumption of a regional water network system during the construction and operation phases. Unlike multi-objective optimization models, this invention prioritizes economic conservation, maximizing energy savings and environmental protection. Furthermore, this model can provide different optimal solutions based on carbon emission requirements, thus better assisting decision-makers in achieving strategic goals of energy conservation, emission reduction, and low-carbon emissions reduction. Attached Figure Description

[0081] Figure 1 A simplified diagram of the simulated case area;

[0082] Figure 2 A schematic diagram of the triangular fuzzy function for user load;

[0083] Figure 3 This is a schematic diagram of the process of obtaining the optimal water pipeline network planning scheme based on bi-level programming and graph theory algorithms in this invention;

[0084] Figure 4 A diagram illustrating the confidence levels of constraints under different carbon emission requirements is presented.

[0085] Figure 5 The diagram shows the corresponding user partition and the optimal pipeline laying scheme. Detailed Implementation

[0086] The present invention will be further defined below with reference to the accompanying drawings and embodiments, but is not limited thereto.

[0087] Example 1

[0088] A water pipeline network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling effect includes the following steps:

[0089] Step 1: Measure and predict data, specifically including:

[0090] Create a simplified map of the area;

[0091] The daily water demand of users is predicted based on their historical water consumption, and represented by a triangular fuzzy membership function.

[0092] Based on the height of the control point above the water surface of the pump station's suction well, the minimum service head required at the control point, and the total head loss from the control point to the pump station, estimate the power consumption of each candidate pipeline segment.

[0093] Measure the required pipe length for laying pipes on each candidate road segment;

[0094] Step 2: Establish a model, which includes an upper-level objective function that minimizes the total annual cost and a lower-level objective function that minimizes the daily electricity consumption of the pipeline network;

[0095] Step 3: Based on bi-level programming and graph theory algorithms, find the optimal value of the model established in Step 2 and obtain the optimal water pipe network planning scheme.

[0096] Example 2

[0097] The water pipeline network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in Example 1 differs in that:

[0098] In step 1, a simplified map of the region is created, such as... Figure 1 As shown, this specifically refers to simplifying candidate water plant locations, user locations, and transportation hub locations into solid squares, solid circles, and dots, respectively. Users are categorized into six types: energy stations, commercial office buildings, residential buildings, hotels, educational buildings, and government buildings, with simple distinctions made. Candidate roads for pipeline laying are simplified into lines.

[0099] In step 1, such as Figure 2 As shown, the membership function μ q The mathematical expression for (x) is shown in equation (Ⅰ):

[0100]

[0101] In equation (Ⅰ), q p The minimum possible water demand value for the user, expressed in meters (m). 3 ;q o The maximum possible water demand value for the user, in meters. 3 ;q m The water demand value for the user with the highest probability, in meters. 3 ;q p and q o The corresponding membership degree is 0, q m The corresponding membership degree is 1.

[0102] In step 1, the power consumption of each candidate pipeline segment is estimated based on the height Z (m) of the control point above the water surface of the pump station's suction well, the minimum service head H (m) required for the control point, and the total head loss h (m) from the control point to the pump station. The specific estimation method is as follows:

[0103] Pipeline e i Electricity consumption per cubic meter of water transport ρ is the density of water, which is 1000 kg / m³. 3 g is the acceleration due to gravity, which is 9.81 m / s². 2 η represents the overall efficiency of the pumping station; Z i pipe segment e i The height difference between the start and end points, H i pipe segment e i Minimum service head, h i pipe segment e i Head loss.

[0104] Example 3

[0105] The water pipeline network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in Example 2 differs in that:

[0106] In step 2, the mathematical expression of the upper-level objective function that minimizes the total annual cost is shown in equation (II):

[0107] TC = WPC + PNC (II)

[0108] In formula (II), TC refers to the annual total cost, WPC refers to the annual cost of the water plant, and PNC refers to the annual cost of the pipeline network.

[0109]

[0110]

[0111] in, WP (representing the water plant) j The minimum possible daily load value is determined by the minimum possible user water demand value. Summation calculation, i.e. WP (representing the water plant) j The daily load value with the highest probability is determined by the water demand value of the user with the highest probability. Summation calculation, i.e. WP (representing the water plant) j The maximum value that may occur in the daily load is the maximum value among the possible user water demand values. Summation calculation, i.e. m represents the total number of water plants designed; f(q) represents the construction cost of the water plants, through... In the calculation, SMP is the estimated parameter; T1 represents the payback period for the water plant construction investment, in years; PRW j Indicates water plant WP j The production cost per cubic meter of water; IPC represents the investment cost of pipeline construction, mainly the cost of pipeline construction, in yuan; T2 represents the payback period of pipeline construction investment; Y1 represents the annual depreciation and overhaul cost of pipeline, in yuan / year, calculated according to a fixed ratio of pipeline construction investment cost; Y2 represents the annual operating cost, in yuan / year, mainly calculating the operating electricity cost.

[0112] The calculation methods for IPC, Y1, and Y2 are as follows:

[0113] IPC=∑ i∈WN (a+b×(D i ) α )×l i ;

[0114]

[0115]

[0116] Where WN represents the set of all selected pipes; a, b, and α represent the statistical parameters of the pipe unit length cost formula; D i Indicates the diameter of pipe i; l i The value indicates the pipeline length; p represents the annual depreciation and overhaul rate of the pipeline network, in percentage; hp represents the maximum head of the pumping station; pre represents the electricity price, in yuan / kWh.

[0117] Pipe diameter D i The calculation method is as follows:

[0118]

[0119] Where v represents the design flow rate; q i The load of pipe i is calculated from the sum of the loads of the downstream users of the pipe. The user load is the user's water demand.

[0120] The mathematical expression for the lower-level objective function of minimizing the daily power consumption of the pipeline network is shown in equation (Ⅲ):

[0121] TE = ECN + EWP + EWD (Ⅲ)

[0122] In formula (Ⅲ), TE refers to the daily electricity consumption of the water pipeline network; ECN refers to the electricity consumption for pipeline production; EWP refers to the electricity consumption for water production; and EWD refers to the electricity consumption for water transmission in the pipeline network.

[0123] The calculations for ECN, EWP, and EWD are as follows:

[0124]

[0125]

[0126]

[0127] Where EFC is the pipeline energy consumption coefficient, the unit is kWh / kg; PG i For pipe weight; EQW j The electricity consumption per cubic meter of water produced by the water plant;

[0128] Pipe weight PG i The calculation method is as follows:

[0129] PG i =π×(R) i 2 -r i 2 )×l i ×RHP=π×(T 2 +D i ×T)×l i ×RHP;

[0130] Among them, R i Represents the outer diameter of the pipe; r i RHP represents the pipe's inner diameter; T represents the pipe material density; T represents the pipe thickness, set to a fixed value.

[0131] The constraints for the upper-level objective function that minimizes the total annual cost and the lower-level objective function that minimizes the daily electricity consumption of the pipeline network include the following:

[0132] a. Each user is supplied with water by only one water plant;

[0133] b. Based on the Mixed possibilistic-flexible robust programming model proposed by Pishvaee et al., the carbon emission limit constraints are calculated as follows: Wherein, CEF is the carbon emission factor for electricity production;

[0134] The confidence level for carbon emission constraints on water supply is at least 50%, i.e., α ≥ 0.5;

[0135] c. Satisfaction level cannot be less than 0 or greater than 1; upper-level decision-makers can determine the lower limit of satisfaction level based on their needs, with the lower limit of satisfaction level assumed to be MS = 0.3, while lower-level decision-makers do not have this authority.

[0136] The calculation method for the overall satisfaction index λ between decision-makers at both upper and lower levels is as follows:

[0137] λ=mon{A,B};0.3≤A≤1;0≤B≤1;

[0138]

[0139]

[0140] in, The minimum economic cost representing only the economic objective is calculated using the A Graph Theory-Based Method proposed by Chen and Zhu. The energy target value f′ corresponding to the minimum economic cost is calculated based on the optimal solution. L Then f′ L The highest energy consumption acceptable to lower-level decision-makers; This refers to the minimum energy consumption required for the energy target, and is based on... The optimal solution is used to determine the highest acceptable economic cost f′ for the upper-level decision-maker. U .

[0141] Example 4

[0142] The water pipeline network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in Example 3 differs in that:

[0143] In step 3, based on bilevel programming and graph theory algorithms, the optimal value of the model established in step 2 is obtained, and the optimal water pipe network planning scheme is acquired, such as... Figure 3 As shown, the specific computer simulation implementation steps include:

[0144] (1) Using the A Graph Theory-Based Method proposed by Chen and Zhu to calculate the optimal solution and optimal value when only the upper-level objective function is considered. The optimal solution considering only the upper-level objective function is the pipeline layout, user zoning, and water plant construction scheme adopted to minimize the total annual cost; optimal value. That is, the minimum total annual cost; and the corresponding lower-level objective value f′ is obtained using the optimal solution. L The lower-level target value f′ L That is, the daily power consumption of the pipeline network;

[0145] Calculate the optimal solution and optimal value when only the lower-level function is considered. The optimal solution considering only the lower-level function refers to the pipeline layout, user zoning, and water plant construction scheme adopted to minimize daily power consumption; optimal value. That is, the minimum daily power consumption of the pipeline network, and the corresponding upper-level target value f′ is obtained using the optimal solution. U Target value f′ U That is, the total annual cost;

[0146] (2) The adjacency matrix and membership function of the simplified input region graph. p q m q o Three values, the candidate pipe segment water transport power consumption estimation matrix EC, and the candidate pipe segment length matrix;

[0147] An adjacency matrix indicates the possibility of laying a pipeline. For example, suppose there are three points A, B, and C in a simplified region map. If only the area between points A and B, and the area between points B and C, has the possibility of laying a pipeline, then the adjacency matrix is ​​a 3x3 matrix, representing rows A, B, and C, and columns A, B, and C, respectively. Based on the aforementioned conditions, we assign a value of 1 to the first row, second column, second row, first column, second row, third column, and third row, second column, and assign a value of 0 to the other positions.

[0148] Candidate pipeline segment water conveyance power consumption estimation matrix EC: pipeline segment e i The estimated power consumption for water transmission is e for the pipeline. i Electricity consumption per cubic meter of water transport E i Multiply by the pipe segment distance LH i Assuming pipe e i With v m Starting from v n If the endpoint is [equation], then the m-row, n-column estimation matrix EC for water transport power consumption of the candidate pipeline segment is E[equation]. i ×LH i Measurements need to be taken based on the actual situation.

[0149] The candidate pipe segment length matrix LH is a matrix composed of the straight-line distances from the start point to the end point of the pipe segment; the matrix has m rows and n columns, recording the lengths starting with v. m Starting from v n Pipeline e as the endpoint i The straight-line distance. Measurements need to be taken based on the actual situation.

[0150] Input refers to the initial values ​​entered into the computer program, which will be used in the calculation of the model's target values ​​in the simulation steps below.

[0151] (3) Enumerate all possible combinations of water plants: by SK K N The set records all possible combinations including K candidate water plants, N = 1, 2, ..., K. N ;

[0152] For each combination, perform a search. After all combinations are searched (K>m), the search ends; otherwise, proceed to step (4) to continue the search. The search process is as follows:

[0153] (4) The shortest water transmission distance between the user and all water plants is calculated by the Floyd algorithm and recorded as matrix SW;

[0154] (5) Assign all users to the water plants in the search combination according to matrix SW, and record the assignment results into matrix SPW;

[0155] (6) Calculate the minimum total load pipeline laying scheme W using the Improved Prim algorithm in the A Graph Theory-Based Method proposed by Chen and Zhu. r (This is obtained through computer simulation using the Improved Prim algorithm); according to the different zones of the water supply plant, the water supply plant, the users supplied by that plant, and the laying path of the water transmission network during the water supply process are recorded in zone P. r middle;

[0156] (7) Calculate the carbon emission constraint confidence level α. If α≤0.5, mark the scheme as infeasible, flag=1, and set N to increase by 1, λ=0, flag=0, and return to step (4) to retrieve the next water plant combination scheme; if α>0.5, proceed to step (8);

[0157] In step (7), the calculation process of the carbon emission constraint confidence level α is as follows:

[0158] Upper limit of carbon emission constraints Calculate the confidence level α; expanding the equation yields... in

[0159] Figure 4 This demonstrates the confidence levels of constraints under different carbon emission requirements. Figure 4 It is evident that under stringent carbon emission requirements, there is a possibility of exceeding carbon emission limits in order to improve decision-makers' satisfaction and ensure water supply. Taking a carbon emission constraint of 32 tons and a two-layer objective function as an example, candidate water plant A and candidate water plant C will be put into construction. Figure 5 It displays the corresponding user partitions and the optimal pipeline laying scheme.

[0160] (8) Calculate the target values ​​f for the upper and lower layers corresponding to the calculation scheme. U =TC,f L =TE; Overall satisfaction of decision-makers at both upper and lower levels λ = min{A,B};

[0161] Table 1 shows the comparison results of the overall satisfaction of this model and the single-objective model under different carbon emission constraints.

[0162] Table 1

[0163] <![CDATA[40t CO2e]]> <![CDATA[36t CO2e]]> <![CDATA[32t CO2e]]> <![CDATA[28t CO2e]]> Single target 0 0 0 1 dual-layer target 0.3936 0.5248 0.8525 1

[0164] As shown in Table 1, this model can effectively improve the overall satisfaction of decision-makers, especially when carbon emission constraints are not strict.

[0165] (9) If the overall satisfaction is negative, the scheme is marked as infeasible, flag=1, and N is increased by 1, λ=0, flag=0, and return to step (4) to retrieve the next water plant combination scheme; if λ is positive, proceed to step (10);

[0166] (10) Determine whether the current solution is the optimal solution, i.e., the solution corresponding to the maximum value of the overall satisfaction λ. If so, record the current solution f. U f L , λ and P r Otherwise, it will not be recorded;

[0167] (11) Set N to increase by 1, λ = 0, flag = 0, and return to step (4) to retrieve the next water plant combination scheme.

[0168] Table 2 shows the optimal water plant site selection and construction plan under different carbon emission constraints.

[0169] Table 2

[0170]

[0171] Decision-makers should choose the optimal solution based on different emission reduction requirements.

[0172] Example 5

[0173] A water network planning system based on bi-level programming and graph theory algorithm under water-energy-carbon coupling effect is used to implement the water network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling effect as described in any of Examples 1-4. The system includes a data measurement and prediction unit, a model building unit, and an optimal water network planning scheme acquisition unit.

[0174] The data measurement and prediction unit is used to perform step 1; the model building unit is used to perform step 2; and the optimal water pipeline network planning scheme finding unit is used to perform step 3.

Claims

1. A water network planning method based on bi-level programming and graph theory algorithms under water-energy-carbon coupling, characterized in that, The steps include the following: Step 1: Measure and predict data, specifically including: Create a simplified map of the area; The daily water demand of users is predicted based on their historical water consumption, and represented by a triangular fuzzy membership function. Based on the height of the control point above the water surface of the pump station's suction well, the minimum service head required at the control point, and the total head loss from the control point to the pump station, estimate the power consumption of each candidate pipeline segment. Measure the required pipe length for laying pipes on each candidate road segment; Step 2: Establish a model, which includes an upper-level objective function that minimizes the total annual cost and a lower-level objective function that minimizes the daily electricity consumption of the pipeline network; Step 3: Based on bilevel programming and graph theory algorithms, find the optimal value of the model established in Step 2 to obtain the optimal water pipeline network planning scheme; the specific implementation steps include: (1) Calculate the optimal solution and optimal value when only considering the upper-level objective function. The optimal solution considering only the upper-level objective function is the pipeline layout, user zoning, and water plant construction scheme adopted to minimize the total annual cost; the optimal value That is, the minimum total annual cost; and the corresponding target value of the next lower level is obtained using the optimal solution. Lower-level target value That is, the daily power consumption of the pipeline network; Calculate the optimal solution and optimal value when only the lower-level function is considered. The optimal solution considering only the lower-level function is the pipeline layout, user zoning, and water plant construction scheme adopted to minimize daily power consumption; optimal value That is, the minimum daily power consumption of the pipeline network, and the corresponding upper-level target value is obtained using the optimal solution. Target value That is, the total annual cost; (2) In the adjacency matrix and membership function of the simplified graph of the input region , , Three values, candidate pipeline segment water conveyance power consumption estimation matrix Candidate pipe segment length matrix; An adjacency matrix is ​​a matrix that indicates whether a pipeline can be laid. Candidate pipe segment length matrix It refers to the matrix formed by the straight-line distances from the beginning to the end of a pipe segment; (3) Enumerate all possible combinations of water plants: by The set records all including Possible combinations of candidate water plants, ; For each combination, perform a search. After all combinations are searched, the search ends; otherwise, proceed to step (4) to continue the search. The search process is as follows: (4) The shortest water transmission distance from the user to all water plants is calculated using the Floyd algorithm and recorded as a matrix. ; (5) Based on the matrix Assign all users to the water plants in the currently searched combination, and record the assignment results in the matrix. ; (6) Calculate the pipeline laying scheme with minimum total load using the Improved Prim algorithm. According to the different zones of the water supply plants, the water supply plants, the users supplied by the plants, and the laying path of the water transmission pipeline network during the water supply process are recorded in the zone. middle; (7) Calculate the confidence level of carbon emission constraints ,like If the value is 0.5, the solution is marked as infeasible. and set Increase by 1, , Return to step (4) to retrieve the next water plant combination scheme; if Then proceed to step (8); (8) Calculate the target values ​​of the upper and lower layers corresponding to the calculation scheme. , ; ; Overall satisfaction of decision-makers at both upper and lower levels ; (9) If the overall satisfaction is A negative value marks the solution as infeasible. and set Increase by 1, , Return to step (4) to retrieve the next water plant combination scheme; if If the value is positive, proceed to step (10); (10) Determine whether the current solution is the optimal solution, i.e., overall satisfaction. Find the solution corresponding to the maximum value; if so, record the current solution. , , and Otherwise, it will not be recorded; (11) Settings Increase by 1, , Return to step (4) to retrieve the next water plant combination scheme; In step (7), the confidence level of carbon emission constraints The calculation process is as follows: Upper limit of carbon emission constraints Calculate confidence level Expanding the equation, we get ,in ; .

2. The water network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in claim 1, characterized in that, Step 1, creating a simplified regional map, specifically refers to: Candidate water plant locations, user locations, and transportation hub locations are simplified to solid squares, solid circles, and dots, respectively. Among them, users are divided into 6 categories: energy stations, commercial office buildings, residential buildings, hotels, teaching buildings, and government buildings, and are simply distinguished. Candidate roads for pipeline laying are simplified to lines.

3. The water network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in claim 1, characterized in that, In step 1, the membership function The mathematical expression for it is shown in equation (Ⅰ): （Ⅰ） In formula (Ⅰ), The minimum possible water demand value for the user, in units of The maximum possible water demand value for the user, in units of and The corresponding membership degree is 0. The corresponding membership degree is 1.

4. The water network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in claim 1, characterized in that, In step 1, the height of the control point above the water surface of the pump station's suction well is determined. ( Minimum service head required for control point ( Total head loss from the control point to the pumping station ( Estimate the power consumption of each candidate pipeline segment; the specific estimation method is as follows: pipeline Electricity consumption per cubic meter of water transport ; The density of water is 1000. ; The acceleration due to gravity is 9.

81. ; To improve the overall efficiency of the pumping station; finger pipe section The height difference between the start and end points finger pipe section Minimum service head finger pipe section Head loss.

5. A water network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in claim 1, characterized in that, In step 2, the mathematical expression of the upper-level objective function that minimizes the total annual cost is shown in equation (II): = + （Ⅱ） In formula (II), This refers to the discounted value of total annual expenses. This refers to the converted value of the water plant's annual costs. This refers to the converted value of the annual pipeline cost; ; ; in, Representing the water plant The minimum possible daily load is calculated by summing the minimum possible user water demand values, i.e. ; Representing the water plant The daily load value with the highest probability is calculated by summing the water demand values ​​of the users with the highest probability. ; Representing the water plant The maximum possible daily load is calculated by summing the maximum possible user water demand values, i.e. ; Indicates the total number of designed water plants; Indicates the construction cost of the water plant, through calculate, These are estimated parameters; This indicates the repayment period for the water plant construction investment, expressed in years. Indicates water plant Production cost per cubic meter of water; This indicates the investment cost for pipeline construction, expressed in yuan. Indicates the repayment period for pipeline construction investment; This represents the annual depreciation and major repair costs of the pipeline network, expressed in yuan / year, and is calculated as a fixed percentage of the pipeline network construction investment cost. This indicates the annual operating cost, expressed in yuan per year.

6. A water network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in claim 1, characterized in that, In step 2, the mathematical expression of the lower-level objective function for minimizing the daily power consumption of the pipeline network is shown in equation (Ⅲ): = + + （Ⅲ） In formula (Ⅲ), This refers to the daily electricity consumption of the water pipe network; This refers to the electricity consumption during pipeline production; This refers to the electricity consumption in water production; This refers to the electricity consumption for water transmission through the pipeline network; , , The calculation is as follows: ； ； ； in, This is the pipeline energy consumption coefficient, in units of... ; This refers to the weight of the pipe. For water plant The electricity consumption for producing one cubic meter of water.

7. A water network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in claim 1, characterized in that, In step 2, the constraints of the upper-level objective function for minimizing the total annual cost and the lower-level objective function for minimizing the daily electricity consumption of the pipeline network include the following: a. Each user is supplied with water by only one water plant; b. Carbon emission limits are calculated as follows: ;in, Carbon emission coefficient for electricity production; The confidence level for carbon emission constraints on water supply is at least 50%, that is... ; c. Satisfaction levels must be neither less than 0 nor greater than 1; upper-level decision-makers have a lower limit on satisfaction levels. Decision-making power; Overall satisfaction index of decision-makers at both upper and lower levels The calculation method is as follows: ； ； ； ； ； in, Representing the minimum economic cost when only an economic objective is achieved, according to The optimal solution is calculated to obtain the energy target value that minimizes economic costs. ,but The highest energy consumption acceptable to lower-level decision-makers; This refers to the minimum energy consumption required for the energy target, and is based on... The optimal solution is used to determine the highest acceptable economic cost for the decision-maker at the higher level. .

8. A water network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in claim 1, characterized in that, =0.3。 9. A water network planning system based on bi-level programming and graph theory algorithms under water-energy-carbon coupling, characterized in that, A water pipeline network planning method based on bi-level programming and graph theory algorithm under water-energy-carbon coupling as described in any one of claims 1-7 is provided, comprising a data measurement and prediction unit, a model building unit, and an optimal water pipeline network planning scheme selection unit; the data measurement and prediction unit is used to perform step 1; the model building unit is used to perform step 2; and the optimal water pipeline network planning scheme selection unit is used to perform step 3.

Images