Urban multi-water-source joint scheduling method based on discrete scene set and robustness decision

Abstract

The invention discloses an urban multi-water-source joint scheduling method based on a discrete scene set and a robustness decision, and the method comprises the steps: researching the characteristics of multiple uncertainty factors of urban multi-source water supply, and generating a multiple uncertainty scheduling scene set through a Latin hypercube sampling method; constructing a water supply city regulation and storage reservoir and external water source joint scheduling model and a water receiving unit internal optimization scheduling model; and constructing a robustness decision-making-based urban annual multi-water-source combined scheduling model. A multi-uncertainty scene set and a robustness measurement method based on an expected benefit index are utilized, the minimization of the annual water level fluctuation risk and the urban water shortage risk of a regulation and storage reservoir is taken as a scheduling target, an urban multi-water-source combined scheduling model based on a robustness decision is constructed, the rationality and the applicability of the robustness scheduling model are verified, and the urban multi-water-source combined scheduling model is established. And analyzing a robust decision scheduling result under an uncertain condition.

Description

Technical Field

[0001] The present invention belongs to the field of water resource optimization configuration, and in particular relates to a new method for joint scheduling of multiple water sources in a city. Background Art

[0002] Optimizing urban water supply scheduling is a crucial measure for ensuring optimal water resource allocation and safeguarding urban water supply security. Influenced by numerous factors during the scheduling process, including the quantity and quality of externally transferred water sources, and urban water demand, optimizing multi-source urban water supply scheduling presents a complex decision-making challenge influenced by multiple uncertainties. Key challenges include comprehensively characterizing these uncertainties and their interrelationships, incorporating urban water supply system risks into optimized scheduling models, and minimizing the overall risk of urban water supply systems under these uncertainties.

[0003] Optimizing urban water supply scheduling is a crucial measure for ensuring optimal water resource allocation and safeguarding urban water supply security. Influenced by numerous factors during the scheduling process, including the quantity and quality of externally transferred water sources, and urban water demand, optimizing multi-source urban water supply scheduling presents a complex decision-making challenge influenced by multiple uncertainties. This invention comprehensively characterizes these uncertainties and their relationships, incorporating urban water supply system risks into the optimized scheduling model to minimize the overall risk of urban water supply systems under uncertain conditions. Summary of the Invention

[0004] In response to the technical problems existing in the above-mentioned prior art, the present invention aims to propose a method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision-making, introduce a robust decision-making method, establish a multi-objective optimization decision-making model based on discrete scenario sets, and study the optimal scheduling method for joint water supply of multiple urban water sources.

[0005] The present invention is achieved by utilizing the following technical solutions:

[0006] The present invention proposes a method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision-making, which realizes the optimal scheduling of joint urban water supply from multiple water sources under multiple uncertain conditions, including the following steps:

[0007] Step 1: Conduct a comprehensive evaluation of the annual raw water diversion source volume and water quality, and analyze the uncertainty factors in the city's annual water consumption forecast. This will identify the characteristics of multiple uncertainties in the city's multi-source water supply, and generate scheduling scenario data for analysis based on multiple uncertainties, including the city's annual water consumption forecast model, the comprehensive water quality evaluation model, the raw water diversion source volume analysis model, and the urban multi-source joint scheduling model.

[0008] Step 2: Construct a joint scheduling model of the water supply city's annual regulating reservoir and external water source, and an optimal scheduling model within the water receiving unit. As a multi-source water supply scheduling model that measures coupling robustness, the joint scheduling model of the water supply city's annual regulating reservoir and external water source further includes two objective functions: minimizing the risk of urban water shortage and minimizing the risk of water level drawdown at the end of the year in the regulating reservoir, and maximizing the water supply capacity of the regulating reservoir.

[0009] The multi-source water supply scheduling model with coupled robustness measurement is to calculate the urban water shortage risk of each scheduling scenario in the set of uncertain scheduling scenarios simulated by Latin hypercube. and the risk of water level drawdown in regulating reservoirs at the end of the year Based on robust decision theory, the robustness measures of various risks under all scheduling scenarios are calculated;

[0010] Consider whether the external water source process meets the water demand of the receiving area and the operation requirements of the regulating reservoir, conduct a multi-source water supply risk analysis for the city within the year, and determine the risk objective function of the water level drawdown of the regulating reservoir at the end of the year The expression is as follows:

[0011]

[0012] Where, The risk of urban water shortage can be converted into the urban water supply guarantee rate. is the risk of water level drawdown at the end of the year in the regulating reservoir, S is the total number of scheduling scenarios, T is the total number of time periods, I is the total number of water supply areas, R is the number of regulating reservoirs, J is the number of regulating reservoirs and external water sources, including R regulating reservoirs and M types of external water sources, WT i，j，t is the actual water supply of the jth water source in the i-th water supply area during the t-th period, DM it is the actual water demand of the water supply area in the t-th period, SCR r，0 is the storage capacity of the rth regulating reservoir at the beginning of the year, SCR r，T is the storage capacity of the rth regulating reservoir at the end of the year;

[0013] Step 3: Analyze the results of the optimized scheduling of urban multi-source water supply. With the goal of minimizing the risk of urban water shortage and the risk of water level drop in the regulating reservoir at the end of the year, construct an urban annual multi-source water supply joint scheduling model based on robust decision-making. The first-level model of the urban annual multi-source water supply joint scheduling model based on robust decision-making is the urban annual regulating reservoir and external water source joint scheduling model, and the second-level model is the optimal scheduling model within the water receiving unit. The input of the urban annual regulating reservoir and external water source joint scheduling model is the cross-basin external water source and the water volume in and out of the regulating reservoir in each period in the future. At the same time, it provides boundary conditions for the optimal scheduling model within the water receiving unit of the second-level model. The output results are two parts: in the set of uncertain scheduling scenarios simulated by Latin hypercube, calculate the urban water shortage risk of each scheduling scenario. and the risk of water level drawdown in regulating reservoirs at the end of the year Based on robust decision theory, the robustness measures of various risks under all scheduling scenarios are calculated;

[0014] Using multiple uncertainty scenario sets and a robustness measurement method based on expected benefit indicators, with the minimization of the risk of water level drawdown in the regulating reservoir and the risk of urban water shortage within the year as the scheduling objective, the rationality and applicability of the robust scheduling model were verified using an urban multi-water source joint scheduling model based on robust decision-making, and the robust decision-making scheduling results under uncertainty were analyzed.

[0015] Step 4: Considering three uncertain factors, including the uncertainty of the water volume of external water sources, the uncertainty of the water quality of external water sources, and the uncertainty of urban water demand, analyze the response relationship between the scheduling target and the scheduling scenario characteristics, and evaluate the impact of the urban annual multi-source water supply joint scheduling model based on robust decision-making on urban water supply scheduling.

[0016] Joint scheduling model of water supply city regulating reservoirs and external water sources and optimal scheduling model within water receiving units

[0017] In some embodiments, the multiple uncertainty analysis of the annual water quantity, water quality, and urban water demand of the water source transferred from outside the city in step 1 further includes the following processing:

[0018] Through the upper and lower boundaries of the confidence interval and the deterministic prediction value, the prediction interval under different confidence levels is determined, and the uncertainty factors faced by the urban water supply process are analyzed and quantified, as shown in the following formula:

[0019]

[0020] Where, represents the deterministic prediction value, β represents the prediction error with a confidence interval, and β L represents the random variable of the forecast error of external water source water volume, β Erepresents the random variable of the water quality prediction error of the external water source, β c represents the random variable of urban water consumption forecast error.

[0021] In some embodiments, the uncertainty factors include:

[0022] The urban annual water consumption prediction model is used to predict the monthly water consumption of urban water plants within a year, as shown in the following formula:

[0023] y pro (t) = g pro (t)+s pro (t)+h pro (t)+ε pro (t)

[0024] Where y pro (t) represents the time series observation value at time t, g pro (t) represents the long-term trend term, S pro (t) represents the periodic term, h pro (t) represents the impact of holiday effects or other special events, ε pro (t) represents the error term;

[0025] The comprehensive evaluation of the water quality of the external water source is to calculate the probability corresponding to the water quality level in each month, as shown in the following formula:

[0026]

[0027] In the formula, by lev，inv Indicates water quality type, lev indicates water quality standard level, ind indicates water quality index, bx ind Indicates the water quality index value, P Bayes (by lev，ind ) represents the prior probability. In the absence of monthly water quality information, the probabilities of different water quality types are equal. P Bayes (bx ind |by lev，ind ) represents the water quality index value bx ind Belong to the water quality type of this indicator lev，ind The possibility of P Bayes (by lev，ind |bx ind ) represents the posterior probability, indicating that the index value bx of the representative month is known ind Under certain conditions, the probability that the water quality of this indicator belongs to the lev level;

[0028] The raw water diversion volume analysis model is used to determine the volume of water that can be diverted at different water inflow frequencies and the volume of water that can be diverted over many years based on existing relevant program results and historical measured data.

[0029] The external water source and water use scenario generation model is used for the joint scheduling of multiple water sources in the city. The model includes a first-level urban external water source and regulating reservoir joint scheduling model, which rationally allocates the water inflow and outflow of cross-basin external water sources and regulating reservoirs; and a second-level water receiving unit optimization scheduling model, which allocates the water inflow and outflow of cross-basin external water sources and regulating reservoirs to different raw water plants for redistribution.

[0030] In some embodiments, the method also includes an analysis model for the amount of raw water that can be diverted externally, which takes the results of existing relevant plans and historical measured data as input to determine the amount of water that can be diverted externally at different water inflow frequencies and the amount of water that can be diverted externally measured over many years as output.

[0031] In some embodiments, the constraints of the joint scheduling model of the water supply city regulating reservoir and external water source in step 2 include:

[0032] i. The water balance constraint of external water source is expressed as follows:

[0033]

[0034] Where, is the total amount of externally diverted water from type m of externally diverted water sources in period t, is the water consumption of the raw water plant from the mth type of external water source in period t, is the reservoir inflow of type m external water source in period t, is the ecological water replenishment of the river from the m-type external water source in period t;

[0035] ii. The water balance constraint of the regulating reservoir is expressed as follows:

[0036]

[0037] Where, is the water storage capacity of the rth reservoir at the end of the t-1 period, is the water storage capacity of the rth reservoir at the end of the tth period, is the inflow of the rth reservoir in the tth period, is the water supply from the rth reservoir to the raw water plant in the tth period, The leakage of the rth reservoir in the i-th period, Evaporation of the rth reservoir in period i;

[0038] iii. Water transmission loss constraint, expressed as follows:

[0039]

[0040] Where, and The water loss coefficient for water transfer from regulating reservoirs and external water sources to urban water users;

[0041] iv. Water supply guarantee rate constraint, expressed as follows:

[0042] P urca1 ≤P urob ≤P urca2

[0043] Where, P urob is the calculated urban water supply guarantee rate, P urca1 and P urca2 To ensure the lower and upper limits of the rate for the set urban water supply targets;

[0044] v. Reservoir capacity constraint, expressed as follows:

[0045]

[0046] Where, is the lower limit of the storage capacity of the rth reservoir in the tth period. The lower limit of the storage capacity of the reservoir is the dead storage capacity or the reduced storage capacity. is the upper limit of the water storage capacity of the rth reservoir in the tth period. The upper limit of the water storage capacity of the reservoir is the storage capacity corresponding to the flood limit water level during the flood season and the storage capacity corresponding to the normal water level during the non-flood season;

[0047] vi. Pipeline water delivery capacity constraint, expressed as follows:

[0048] WPE n,t ≤DE n

[0049] Where WPE n，t is the water delivery of the nth water pipeline in the tth period, DE n is the actual maximum water delivery capacity of the nth water pipeline;

[0050] vii. Raw water quality constraints: When the water quality index does not meet the surface water Class III standard and raw water emergency pretreatment is not considered, the amount of raw water that can be diverted is set to 0;

[0051] viii. Non-negative constraints on variables: The model must satisfy the non-negative constraints on decision variables.

[0052] In some embodiments, the optimization scheduling objective function in the water receiving unit in step 2 is expressed as follows:

[0053]

[0054] Where: is the comprehensive relative water shortage of the raw water plant, DM p，t is the water demand of the pth water plant during period t; WTI j，p，tis the water supply from the jth water source to the pth water plant in the tth period, T is the period number, P is the raw water plant number, α p is the importance parameter of the pth water plant;

[0055] The optimal scheduling constraint function within the water receiving unit is defined as follows:

[0056] i. Constraints on available water supply from water sources, expressed as follows:

[0057] The total amount of water supplied by the source to each water user shall not exceed the water supply of the source. The water supply constraint of this study refers to the water diversion from the Yangtze River, the Luan River and the storage capacity of the regulating reservoir:

[0058] WTI j,p,t ≤SC j,p

[0059] Where: SC j，p The maximum water supply capacity of the jth type of water source to the pth water plant;

[0060] ii. The water supply capacity constraint of the pipeline project is expressed as follows:

[0061] PS k,t ≤GC k,t

[0062] Where, PS k，t is the water delivery of the kth water delivery channel in the tth period, GC k,t is the actual maximum water delivery capacity of the kth water pipeline;

[0063] iii. Constraints on the water purification capacity of the water plant during the period, , are expressed as follows:

[0064] DM p,t ≤PC p,t

[0065] Where: PC p,t is the maximum water purification capacity of the water plant p at time t;

[0066] iv. Raw water quality constraint, expressed as follows:

[0067] When the water quality index does not meet the surface water Class III standard and the raw water emergency pretreatment is not considered, the amount of raw water that can be diverted externally is set to 0.

[0068] v. The water supply constraint is a decision variable that satisfies the non-negative constraint. The expression is as follows:

[0069] WTI j,p,t >0.

[0070] In some embodiments, the method further includes generating a scheduling scenario set of multiple uncertainties for the urban multi-source water supply by a Latin hypercube sampling method; the steps are as follows:

[0071] For N uncertainty factor variables corresponding to N dimensions, each uncertainty factor variable is divided into M intervals with equal probability; a random uncertainty factor variable value is selected in each interval of each dimension, ensuring that each interval has one and only one sample value; the above steps are repeated until the required number of sample points are generated to cover the entire uncertainty factor parameter space.

[0072] In some embodiments, the joint scheduling model of the city's annual regulating reservoirs and external water sources includes two objective functions: minimizing the risk of urban water shortage and minimizing the risk of water level drawdown in the regulating reservoir at the end of the year. Under the constraints, the allocation of external water transfer volume and the inflow and outflow of the regulating reservoir are output. According to the available water volume determined by the joint scheduling model of the city's regulating reservoirs and external water sources, the water delivery volume of the raw water pipeline in each future time period is determined with the water demand of each user of the water receiving unit, the water delivery capacity of the pipeline and other factors as constraints, so as to realize the allocation of various water sources among different users in each water receiving unit.

[0073] Compared with the prior art, the advantages and positive technical effects achieved by the present invention are as follows:

[0074] 1) Comprehensively characterize various uncertainty factors and their correlations, incorporate urban water supply system risks into the optimization scheduling model, and minimize the comprehensive risks of urban water supply systems under uncertain conditions;

[0075] 2) Introducing robust decision-making methods, establishing a multi-objective optimization decision-making model based on discrete scenario sets, and studying methods for optimizing the coordinated scheduling of urban water supply from multiple water sources. Proposing a multi-uncertainty analysis method for annual water volume and quality of externally transferred water sources, as well as urban water demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a flow chart of the urban multi-water source joint scheduling method based on discrete scenario sets and robust decision-making of the present invention;

[0077] Figure 2 A schematic diagram of the structure of a multi-source water supply scheduling model for coupling robustness measurement in step 2;

[0078] Figure 3 This is a schematic diagram of the calculation results of the urban annual multi-source water supply joint scheduling model based on robust decision-making in step three;

[0079] Figure 4 This is a schematic diagram of the solution results of the urban annual multi-source water supply joint scheduling model based on robust decision-making in step three, taking into account the risk of water quality degradation of external water sources;

[0080] Figure 5 A schematic diagram of the scheduling results of the urban multi-source water supply joint scheduling model based on robust decision-making in step 3, corresponding to the amount of externally transferred water sources considering the risk of water quality degradation of externally transferred water sources;

[0081] Figure 6 It is the probability density and cumulative distribution function diagram of scheduling target;

[0082] Figure 7 Schematic diagram of the impact of uncertain factors on water supply scheduling. DETAILED DESCRIPTION

[0083] The present invention will be further described below with reference to the accompanying drawings and examples.

[0084] like Figure 1 As shown in FIG, the overall process of the urban multi-water source joint scheduling method based on the robust decision of discrete scenario sets of the present invention is applicable to application scenarios with both raw water and external water sources. The specific steps are as follows:

[0085] Step 1: Generate scheduling scenario data to be analyzed based on multiple uncertain factors, including the city's annual water consumption prediction model, water quality comprehensive evaluation model, raw water diversion source analysis model, and diversion water source and water use scenario generation model, namely;

[0086] 1-1, the construction of urban annual water consumption prediction model, that is, based on the Prophet time series prediction method, the monthly water volume of the first three years is input and the monthly water volume of the next year is output. The annual water consumption prediction model of the urban water purification plant is constructed. The measured water volume data of the water plant in a certain period of time is collected. The training set and validation set are constructed according to a certain ratio based on the measured data. The hyperparameters of the model are set by the grid search method. For the periodic term of the annual scale, P is set. pro is 365, N p is 10; for the periodic term on a weekly scale, P pro is 7, N p is 3, where P pro is the length of the time series corresponding to the periodic term on an annual scale, N p is the number of cycles corresponding to the annual periodic term; the trend term prediction method is the Logistic curve, and the main hyperparameters of the Prophet model are shown in Table 1. Table 1 shows the main hyperparameters of the Prophet model.

[0087] Table 1

[0088]

[0089] Detailed description of the city's annual water consumption forecast:

[0090] The Prophet time series forecasting method is used to predict the monthly water consumption of urban water plants within a year. The Prophet method uses the theories of empirical mode decomposition (EMD) and generalized additive models (GAM) to decompose time series data into trend terms, cyclic terms, and special terms. Based on this, the time series data is forecasted as shown in the formula.

[0091] y pro (t) = g pro (t)+s pro (t)+h pro (t)+ε pro (t) (1)

[0092] Where y pro (t) represents the time series observation value at time t, g pro (t) represents the long-term trend term, s pro (t) represents the periodic term, h pro (t) represents the impact of holiday effects or other special events, ε pro (t) represents the error term. The Prophet method obtains the forecast results of time series data by modeling and combining these components;

[0093] 1-2. Construct a comprehensive water quality evaluation model for external water sources. This model uses measured data from (year) to (year) and selects five water quality evaluation indicators: pH, dissolved oxygen, permanganate index, ammonia nitrogen, and total phosphorus. Using Bayesian theory, this model is constructed to calculate the probability of water quality levels for each month and analyze and evaluate the water quality characteristics of external water sources. The model specifically includes the following components:

[0094] Using years of measured data and Bayesian theory, we established a comprehensive water quality evaluation model, conducted annual water quality evaluations of externally transferred raw water, and analyzed and calculated the probabilities corresponding to the water quality levels in each month, laying the data foundation for the subsequent generation of a set of scheduling scenarios. The calculation formula is shown as follows:

[0095]

[0096] In the formula, by lev,ind Indicates water quality type, lev indicates water quality standard level, ind indicates water quality index, bx ind Indicates the water quality index value, P Bayes (by lev,ind ) represents the prior probability. In the absence of monthly water quality information, the probabilities of different water quality types are equal. P Bayes (bx ind |by lev，ind ) represents the water quality index value bx ind Belong to the water quality type of this indicator lev，indThe greater the value, the greater the possibility of belonging to a certain water quality level. The normal distribution principle is used to calculate P Bayes (bx ind |by lev，ind ), P Bayes (by lev，ind |bx ind ) represents the posterior probability, indicating that the index value bx of the representative month is known ind Under certain conditions, the probability that the water quality of this indicator belongs to the lev level.

[0097] 1-3. Construction of an analysis model for the amount of raw water that can be diverted externally, which uses the results of existing relevant plans and historical measured data as input, and determines the amount of water that can be diverted externally under different water inflow frequencies and the amount of water that can be diverted externally over many years as output, laying the foundation for generating multiple uncertain scheduling scenarios.

[0098] 1-4, construct an external water source and water use scenario generation model for the joint scheduling of multiple water sources in the city. The model includes the first-level urban external water source and regulating reservoir joint scheduling model, which rationally allocates the cross-basin external water source and the water inflow and outflow of regulating reservoirs; the second-level water receiving unit optimization scheduling model allocates the cross-basin external water source and the water inflow and outflow of regulating reservoirs to different raw water plants for redistribution. For example, the external water source cycle is the entire water conservancy year, that is, from November to October of the following year, and one month is taken as an optimization period; the water conservancy year refers to the annual external water source time. The solution method uses the month as the water allocation period to solve and study the multi-water source joint water supply operation process, and realize the reasonable allocation plan of water resources in time, space, quantity and quality. The specific description of the external water source and water use scenario generation is as follows:

[0099] External Water Sources: Considering the uncertainty of water volume and water quality from external water sources, based on existing plans and measured data from XX to XX, the available water volume for different inflow frequencies and the measured water volume over multiple years were determined. Using the measured data from XX to XX, a comprehensive water quality evaluation model was developed using Bayesian theory. Five water quality evaluation indicators (pH, dissolved oxygen, permanganate index, ammonia nitrogen, and total phosphorus) were selected as water quality evaluation indicators to analyze and evaluate the water quality of external water sources. A set of annual water transfer scenarios was generated using the Latin oversampling multiple uncertainty scheduling scenario generation method.

[0100] Water use scenario generation: Utilizing measured water volume data from a water purification plant from XX to XX, a Prophet time series forecasting method was used to construct an annual water use forecasting model for urban water plants, targeting urban water plants. This model used monthly water volume from the previous three years as input and monthly water volume from the next year as output. This model predicted urban water use from October 2021 to November 2022. A set of annual water use scenarios for different cities was generated using the Latin oversampling method.

[0101] Step 2: Construct a joint scheduling model for the water supply city's annual regulating reservoirs and external water sources and an optimal scheduling model within the water receiving unit. This multi-source water supply scheduling model, which serves as a coupling robustness measure, is specifically described as follows:

[0102] Taking into account the two types of water sources, namely, external water sources and storage water, and on the premise of meeting various constraints such as the water demand of raw water plants and project safety, a joint scheduling model of urban storage reservoirs and external water sources (first-level model) is constructed to allocate cross-basin external water sources and regulate the water inflow and outflow of reservoirs in various time periods in the future, providing boundary conditions for the optimal scheduling model (second-level model) within the water receiving unit.

[0103] 2-1. The detailed description of the joint scheduling model of water storage reservoirs and external water sources in water supply cities within a year is as follows:

[0104] 2-1-1. The objective function of the joint scheduling model of water storage reservoirs and external water sources in water supply cities within a year includes two types of objectives, which are specifically defined as follows:

[0105] 1) Minimum risk of urban water shortage: Minimum risk of urban water shortage can be converted into maximum urban water supply guarantee rate.

[0106] 2) The risk of water level drawdown at the end of the year in regulating and storage reservoirs is minimal and their water supply capacity is maximized: Regulating and storage reservoirs are important regulating and backup water sources for inter-basin water diversion projects. Increasing the water supply capacity of regulating and storage reservoirs can reduce the impact of external water source security issues such as water outages in the main canal and reduced water flow during ice seasons on urban water supply, thereby ensuring stable urban water supply flow. Joint scheduling model of regulating and storage reservoirs and external water sources in water supply cities within the year It is defined as achieving the best decision for robustness, and the expression is as follows:

[0107]

[0108] Where, The risk of urban water shortage can be converted into the urban water supply guarantee rate. is the risk of water level drawdown at the end of the year in the regulating reservoir, S is the total number of scheduling scenarios, T is the total number of time periods, I is the total number of water supply areas, R is the number of regulating reservoirs, J is the number of regulating reservoirs and external water sources, including R regulating reservoirs and M types of external water sources, WT i，j，t is the actual water supply of the jth water source in the i-th water supply area during the t-th period, DM it is the actual water demand of the water supply area in the t-th period, SCR r，0 is the storage capacity of the rth regulating reservoir at the beginning of the year, SCR r，T is the storage capacity of the rth regulating reservoir at the end of the year.

[0109] Furthermore, the uncertainty scheduling scenario set is mainly considered to ensure the lowest risk of urban water shortage and the year-end water level drawdown risk of the regulating reservoir, and to reasonably allocate water resources in time, space, quantity and among users. The multi-source water supply scheduling model structure coupled with robustness measurement is as follows: Figure 2 As shown in the figure, the model mainly consists of two parts: calculating the urban water shortage risk of each scheduling scenario in the uncertain scheduling scenario set simulated by Latin hypercube and the risk of water level drawdown in regulating reservoirs at the end of the year Based on robust decision theory, robustness metrics are calculated for various types of risks under all scheduling scenarios. When calculating various risks under scheduling scenarios, the model's computational complexity increases dramatically as the number of water-receiving units, the number of users, and related indicators increases, potentially leading to the "curse of dimensionality" problem. To address this issue, the urban multi-source water supply joint scheduling model is decomposed into a two-level hierarchical model based on the concept of dimensionality reduction to achieve multi-source joint scheduling and optimal allocation among multiple users. The first level is a joint scheduling model for urban external water sources and regulating reservoirs, which rationally allocates the inflow and outflow of cross-basin external water sources and regulating reservoirs. The second level is an optimized scheduling model within the original water-receiving unit, which redistributes the inflow and outflow of cross-basin external water sources and regulating reservoirs to different raw water plants.

[0110] 2-2, the constraints of the joint scheduling model of urban water storage reservoirs and external water sources for annual water supply are defined as follows:

[0111] i. The water balance constraint of external water source is expressed as follows:

[0112]

[0113] Where, is the total amount of externally diverted water from type m of externally diverted water sources in period t, is the water consumption of the raw water plant from the mth type of external water source in period t, is the reservoir inflow of type m external water source in period t, It is the river ecological water replenishment of type m external water source in period t.

[0114] ii. The water balance constraint of the regulating reservoir is expressed as follows:

[0115]

[0116] Where, is the water storage capacity of the rth reservoir at the end of the t-1 period, is the water storage capacity of the rth reservoir at the end of the tth period, is the inflow of the rth reservoir in the tth period, is the water supply from the rth reservoir to the raw water plant in the tth period, The leakage of the rth reservoir in the i-th period, Evaporation of the rth reservoir in the i-th period.

[0117] iii. Water transmission loss constraint, expressed as follows:

[0118]

[0119] Where, and It is the water loss coefficient from regulating reservoirs and external water sources to urban water users.

[0120] iv. Water supply guarantee rate constraint, expressed as follows:

[0121] P urca1 ≤P urob ≤P urca2 (10)

[0122] Where, P urob is the calculated urban water supply guarantee rate, P urca1 and P urca2 The lower and upper limits of the guaranteed rate are set for the urban water supply targets.

[0123] v. Reservoir capacity constraint, expressed as follows:

[0124]

[0125] Where, is the lower limit of the storage capacity of the rth reservoir in the tth period. The lower limit of the storage capacity of the reservoir is the dead storage capacity or the reduced storage capacity. is the upper limit of the water storage capacity of the rth reservoir in the tth period. The upper limit of the water storage capacity of the reservoir is the storage capacity corresponding to the flood limit water level during the flood season and the storage capacity corresponding to the normal water level during the non-flood season.

[0126] vi. Pipeline water delivery capacity constraint, expressed as follows:

[0127] WPE n,t ≤DE n (10)

[0128] Where WPE n，t is the water delivery of the nth water pipeline in the tth period, DE n is the actual maximum water delivery capacity of the nth water pipeline.

[0129] vii. Raw water quality constraints:

[0130] When the water quality index does not meet the surface water Class III standard and the raw water emergency pretreatment is not considered, the amount of raw water that can be diverted externally is set to 0.

[0131] viii. Non-negative constraints on variables:

[0132] The model must satisfy the non-negativity constraint of decision variables.

[0133] Optimal scheduling objective function within the water receiving unit

[0134]

[0135] Where: is the comprehensive relative water shortage of the raw water plant, DM p，t is the water demand of the pth water plant during period t; WTI j，p，t is the water supply from the jth water source to the pth water plant in the tth period, T is the period number, P is the raw water plant number, α p is the importance parameter of the p-th water plant.

[0136] 2-3, the optimal scheduling constraint function within the water receiving unit is defined as follows:

[0137] i. Constraints on available water supply from water sources, expressed as follows:

[0138] The sum of the water supply from the source to each water user shall not exceed the water supply capacity of the water source. The water supply capacity constraint of this study refers to the water diverted from the Yangtze River, the Luan River and the storage capacity of the regulating reservoir.

[0139] WTI j,p,t ≤SC j,p (12)

[0140] Where: SC j，p The maximum water supply capacity of the jth type of water source to the pth water plant.

[0141] ii. The water supply capacity constraint of the pipeline project is expressed as follows:

[0142] PS k,t ≤GC k,t (13)

[0143] Where, PS k，t is the water delivery of the kth water delivery channel in the tth period, GC k，t is the actual maximum water delivery capacity of the kth water pipeline.

[0144] iii. Constraints on the water purification capacity of the water plant during the period, , are expressed as follows:

[0145] DM p,t ≤PC p,t (14)

[0146] Where: PC p，t is the maximum water purification capacity of the water plant p at time t.

[0147] iv. Raw water quality constraint, expressed as follows:

[0148] When the water quality index does not meet the surface water Class III standard and the raw water emergency pretreatment is not considered, the amount of raw water that can be diverted externally is set to 0.

[0149] v. Non-negative constraint on variables, the expression is as follows:

[0150] The model must satisfy the non-negativity constraint of decision variables.

[0151] WTI j,p,t >0 (15)

[0152] Step 3: Analyze the results of the optimized scheduling of urban multi-source water supply; the details are as follows:

[0153] Focusing on the uncertainty of the amount of raw water that can be diverted from external sources, the water quality of raw water that can be diverted from external sources, and the city's annual water consumption forecast during the scheduling process, a set of multiple uncertain scheduling scenarios is used, with the goal of minimizing the risk of urban water shortage and the risk of water level drop in the regulating reservoir at the end of the year. A city's annual multi-source water supply joint scheduling model based on robust decision-making is constructed to analyze the risk of urban water shortage and the risk of water level drop in the regulating reservoir at the end of the year. The city's annual multi-source water supply joint scheduling model based on robust decision-making of the present invention includes two types of scheduling targets: water supply guarantee rate target and regulating reservoir end-of-year storage capacity reduction target. Priority is given to meeting the city's water supply guarantee rate target, that is, the water demand of the urban water plant is met by discharging water from the regulating reservoir, and the optimal solution set for the city's water supply guarantee rate target is sought. A relatively good solution for the regulating reservoir end-of-year storage capacity reduction target is found in the optimal solution set; priority is given to ensuring the city's water supply guarantee rate target, while taking into account the regulating reservoir end-of-year storage capacity reduction target.

[0154] 3-1. Consider two different water quality risk calculation schemes to build a city multi-source water supply joint scheduling model based on robust decision-making. The model calculation results are as follows: Figure 3 As shown;

[0155] Scheme 1 is a calculation scheme that does not consider the risk of water quality degradation of the water used, and Scheme 2 is a calculation scheme that considers the risk of water quality degradation of the water used. The average urban water supply guarantee rate of Scheme 1 is 99.86%, and the average storage capacity of the regulating reservoir at the end of the year is 793,000 m 3 The average urban water supply guarantee rate of Scheme 2 is 98.98%, and the average storage capacity of the regulating reservoir at the end of the year is 4.426 million m 3 Compared with the case where the risk of water quality decline of external water sources is taken into account, the urban water supply guarantee rate increases by an average of 0.88% when the risk of water quality decline is not taken into account, and the storage capacity of the regulating reservoir increases by an average of 3.633 million m3 at the end of the year. 3 The risk of urban water shortage and the risk of water level drawdown in regulating reservoirs at the end of the year have been significantly reduced.

[0156] The detailed description of the urban multi-source water supply joint scheduling model based on robust decision-making is as follows:

[0157] The first-level model input of the "Urban Multi-Source Water Supply Joint Scheduling Model Based on Robust Decision-Making" is the inter-basin water transfer and reservoir inflow and outflow regulation in each future period. It also provides boundary conditions for the optimal scheduling model (second-level model) within the water receiving unit. The output results are two parts:

[0158] 1) Calculate the urban water shortage risk of each scheduling scenario in the set of uncertain scheduling scenarios simulated by Latin Hypercube and the risk of water level drawdown in regulating reservoirs at the end of the year

[0159] 2) Based on robust decision theory, calculate the robustness metrics of various risks under all scheduling scenarios.

[0160] The joint scheduling model of urban regulating reservoirs and external water sources within the year (the first-level model) includes two goals: to minimize the risk of water shortage for urban users and to minimize the risk of water level drop in the regulating reservoir at the end of the year. Under various constraints, the output of external water volume distribution and the inflow and outflow of regulating reservoirs are determined. According to the various water supply volumes determined by the joint scheduling model of urban regulating reservoirs and external water sources, with the water demand of each user in the water receiving unit and the water delivery capacity of the pipeline as constraints, the water delivery volume of the raw water pipeline in each time period in the future is determined to realize the allocation of various water sources among different users in each water receiving unit. (It is an idea of subdividing the whole into parts);

[0161] 3-2. Consider the risk of water quality degradation of external water sources, as described below:

[0162] When considering the risk of water quality degradation of external water sources, the calculation results of the urban multi-source water supply joint scheduling model based on robust decision-making corresponding to each external water source scenario are as follows: Figure 4 As shown in the figure, the dispatching results corresponding to different external water sources are as follows: Figure 5 Due to the limitation of water source conditions, considering the risk of water quality decline of external water sources, the amount of water transferred from inside and outside the country is more than 900 million m 3 The risk of urban water shortage has been significantly reduced, with the amount of water transferred from inside and outside the city exceeding 1.1 billion m3. 3 The risk of water level drawdown in regulating reservoirs at the end of the year has been significantly reduced.

[0163] Step 4: Evaluate the impact of uncertainties on water supply scheduling; the details are as follows:

[0164] In order to further study the impact of uncertainty factors on water supply scheduling, the model results were evaluated under a set of uncertainty scenarios, and the response relationship between the scheduling objectives of the urban multi-source water supply joint scheduling model within the year based on robust decision-making and the scheduling scenario characteristics was analyzed. In the scheduling model, the key considerations of the uncertainty factors mainly include three types: uncertainty in the water volume of external water sources, uncertainty in the water quality of external water sources, and uncertainty in urban water demand.

[0165] In the urban multi-source water supply joint scheduling model based on robust decision-making, the probability density (PD) and cumulative distribution function (CDF) of the scheduling target are as follows: Figure 6 As shown in the figure, compared with the case without considering the risk of water quality degradation of external water sources, the PD and CDF distribution range of the scheduling target considering the risk of water quality degradation is wider, the scheduling risk of the 10% most unfavorable scenario is significantly increased, and the overall scheduling risk is increased to a certain extent. The water quality of external water sources is an important factor affecting the risk of urban water shortage and the risk of water level drawdown in regulating reservoirs at the end of the year. Water quality improvement measures can help reduce the potential risks of adverse working conditions.

[0166] For the set of uncertain scheduling scenarios, according to the definitions of characteristic indicators of various scheduling scenarios, characteristic indicators of different scheduling scenarios are calculated, including the mean (Q_mean), median (Q_med), peak coefficient (Q_ske), standard deviation (Q_std), skewness coefficient (Q_kurt) of the amount of water transferred from outside in each month of the year, as well as the mean (W_mean) and median (W_med) of water consumption; Pearson correlation analysis is used to calculate the Pearson correlation coefficient between the scheduling target and the scheduling scenario characteristic indicators. The closer the Pearson correlation coefficient is to -1 or 1, the higher the correlation between the scheduling scenario characteristic indicators and the scheduling target. The Pearson correlation between the scheduling scenario characteristic indicators and the scheduling target is calculated, as shown in the following example: Figure 7 shown.

[0167] In step 1, multiple uncertainty analyses are performed on the annual water volume and quality of the externally transferred water sources and the urban water demand. The prediction intervals at different confidence levels are determined by the upper and lower bounds of the confidence intervals and the deterministic prediction values. The uncertainty factors faced by the urban water supply process are analyzed and quantified, as shown in the following formula:

[0168]

[0169] In the formula, ξ represents the deterministic prediction value, β represents the prediction error with a confidence interval, and β L represents the random variable of the forecast error of external water source water volume, β Erepresents the random variable of the water quality prediction error of the external water source, β C represents the random variable of urban water consumption forecast error.

[0170] These include the following three uncertainties:

[0171] Forecast of urban water consumption within the year.

[0172] The Prophet time series forecasting method is used to predict the monthly water consumption of urban water plants within a year. The Prophet method uses the theories of empirical mode decomposition (EMD) and generalized additive models (GAM) to decompose time series data into trend terms, cyclic terms, and special terms. Based on this, the time series data is forecasted as shown in the formula.

[0173] y pro (t) = g pro (t)+s pro (t)+h pro (t)+ε pro (t)

[0174] Where y pro (t) represents the time series observation value at time t, g pro (t) represents the long-term trend term, S pro (t) represents the periodic term, h pro (t) represents the impact of holiday effects or other special events, ε pro (t) represents the error term. The Prophet method obtains the forecast results of time series data by modeling and combining these components.

[0175] a) Analysis of the quality of the raw water transferred from outside.

[0176] Using years of measured data and Bayesian theory, we established a comprehensive water quality evaluation model, conducted annual water quality evaluations of externally transferred raw water, and analyzed and calculated the probabilities corresponding to the water quality levels in each month, laying the data foundation for the subsequent generation of a set of scheduling scenarios. The calculation formula is shown as follows:

[0177]

[0178] In the formula, by lev，ind Indicates water quality type, lev indicates water quality standard level, ind indicates water quality index, bx ind Indicates the water quality index value, P Bayes (by lev，ind ) represents the prior probability. In the absence of monthly water quality information, the probabilities of different water quality types are equal. P Bayes (bx ind |by lev，ind ) represents the water quality index value bx indBelong to the water quality type of this indicator lev，ind The greater the value, the greater the possibility of belonging to a certain water quality level. The normal distribution principle is used to calculate P Bayes (bx ind |by lev，ind ), P Bayes (by lev，ind |bx ind ) represents the posterior probability, indicating that the index value bx of the representative month is known ind Under certain conditions, the probability that the water quality of this indicator belongs to the lev level.

[0179] Analysis of the amount of raw water that can be diverted: Based on the results of existing relevant plans and historical measured data, the amount of water that can be diverted under different water inflow frequencies and the amount of water that can be diverted over many years are determined, laying the foundation for generating multiple uncertain scheduling scenarios.

[0180] Specifically, the present invention further includes generating a set of multiple uncertainty scheduling scenarios through a Latin hypercube sampling method; the steps of the Latin hypercube sampling method are as follows:

[0181] 1) Assume there are N variables (dimensions), and divide each variable into M intervals with equal probability.

[0182] 2) Randomly select a value in each interval of each dimension, ensuring that each interval has one and only one sample value.

[0183] 3) Repeat the above steps until the required number of sample points are generated to cover the entire parameter space.

[0184] The above is a detailed introduction to a method for urban annual multi-water source joint scheduling based on robust decision-making of discrete scenario sets under multiple uncertainties provided by this application. It should be noted that the present invention is not limited to the above-mentioned method steps and calculation process. The above-mentioned specific implementation methods are merely illustrative and non-restrictive. Researchers in this field can, inspired by the present invention, make formal changes and modifications to the present invention without departing from the purpose of the present invention and the scope of protection of the claims, and all of these are protected by the present invention.

[0185] Based on the embodiments of the present invention, all other embodiments and technical replacements and modifications of the embodiments obtained by ordinary technicians in this field without departing from the spirit of the present invention and without making creative work shall fall within the scope of protection of the present invention.

[0186] A multi-uncertainty analysis method for the annual water volume and quality of externally transferred water sources and urban water demand is proposed. The characteristics of the multiple uncertainty factors in urban multi-source water supply are studied, and a set of multi-uncertainty scheduling scenarios is generated using the Latin hypercube sampling method. A model for the annual joint scheduling of urban multi-source water sources based on robust decision-making is constructed. Using a set of multi-uncertainty scenarios and a robustness measurement method based on expected benefit indicators, a model for the annual joint scheduling of urban multi-source water sources based on robust decision-making is constructed, with the scheduling objectives of minimizing the risk of annual water level drawdown in regulating and storage reservoirs and the risk of urban water shortages. The rationality and applicability of the robust scheduling model are verified, and the robust decision-making scheduling results under uncertainty are analyzed.

Claims

1. A method for joint scheduling of urban multi-water sources based on discrete scenario sets and robust decision-making to achieve optimal scheduling of urban multi-water source joint water supply under multiple uncertain conditions, characterized by: The following steps are involved: Step 1: Conduct a comprehensive evaluation of the annual raw water diversion source volume and water quality, and analyze the uncertainty factors in the city's annual water consumption forecast. This will identify the characteristics of multiple uncertainties in the city's multi-source water supply, and generate scheduling scenario data for analysis based on multiple uncertainties, including the city's annual water consumption forecast model, the comprehensive water quality evaluation model, the raw water diversion source volume analysis model, and the urban multi-source joint scheduling model. Step 2: Construct a joint scheduling model of the water supply city's annual regulating reservoir and external water source, and an optimal scheduling model within the water receiving unit. As a multi-source water supply scheduling model that measures coupling robustness, the joint scheduling model of the water supply city's annual regulating reservoir and external water source further includes two objective functions: minimizing the risk of urban water shortage and minimizing the risk of water level drawdown at the end of the year in the regulating reservoir, and maximizing the water supply capacity of the regulating reservoir. The multi-source water supply scheduling model with coupled robustness measurement is to calculate the urban water shortage risk of each scheduling scenario in the set of uncertain scheduling scenarios simulated by Latin hypercube. and the risk of water level drawdown in regulating reservoirs at the end of the year Based on robust decision theory, the robustness measures of various risks under all scheduling scenarios are calculated; Consider whether the external water source process meets the water demand of the receiving area and the operation requirements of the regulating reservoir, conduct a multi-source water supply risk analysis for the city within the year, and determine the risk objective function of the water level drawdown of the regulating reservoir at the end of the year The expression is as follows: Where, The risk of urban water shortage can be converted into the urban water supply guarantee rate. is the risk of water level drawdown at the end of the year in the regulating reservoir, S is the total number of scheduling scenarios, T is the total number of time periods, I is the total number of water supply areas, R is the number of regulating reservoirs, J is the number of regulating reservoirs and external water sources, including R regulating reservoirs and M types of external water sources, WT i，j，t is the actual water supply of the jth water source in the i-th water supply area during the t-th period, DM it is the actual water demand of the water supply area in the t-th period, SCR r，0 is the storage capacity of the rth regulating reservoir at the beginning of the year, SCR r，T is the storage capacity of the rth regulating reservoir at the end of the year; Step 3: Analyze the results of the optimized scheduling of urban multi-source water supply. With the goal of minimizing the risk of urban water shortage and the risk of water level drop in the regulating reservoir at the end of the year, construct an urban annual multi-source water supply joint scheduling model based on robust decision-making. The first-level model of the urban annual multi-source water supply joint scheduling model based on robust decision-making is the urban annual regulating reservoir and external water source joint scheduling model, and the second-level model is the optimal scheduling model within the water receiving unit. The input of the urban annual regulating reservoir and external water source joint scheduling model is the cross-basin external water source and the water volume in and out of the regulating reservoir in each period in the future. At the same time, it provides boundary conditions for the optimal scheduling model within the water receiving unit of the second-level model. The output results are two parts: in the set of uncertain scheduling scenarios simulated by Latin hypercube, calculate the urban water shortage risk of each scheduling scenario. and the risk of water level drawdown in regulating reservoirs at the end of the year Based on robust decision theory, the robustness measures of various risks under all scheduling scenarios are calculated; Using multiple uncertainty scenario sets and a robustness measurement method based on expected benefit indicators, with the minimization of the risk of water level drawdown in the regulating reservoir and the risk of urban water shortage within the year as the scheduling objective, the rationality and applicability of the robust scheduling model were verified using an urban multi-water source joint scheduling model based on robust decision-making, and the robust decision-making scheduling results under uncertainty were analyzed. Step 4: Considering three uncertain factors, including the uncertainty of the water volume of external water sources, the uncertainty of the water quality of external water sources, and the uncertainty of urban water demand, analyze the response relationship between the scheduling target and the scheduling scenario characteristics, and evaluate the impact of the urban annual multi-source water supply joint scheduling model based on robust decision-making on urban water supply scheduling.

2. The method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision-making according to claim 1 is characterized in that: The multiple uncertainty analysis of the annual water quantity, water quality, and urban water demand of the water source transferred from outside the city in step 1 further includes the following processing: Through the upper and lower boundaries of the confidence interval and the deterministic prediction value, the prediction interval under different confidence levels is determined, and the uncertainty factors faced by the urban water supply process are analyzed and quantified, as shown in the following formula: Where, represents the deterministic prediction value, β represents the prediction error with a confidence interval, and β L represents the random variable of the forecast error of external water source water volume, β E represents the random variable of the water quality prediction error of the external water source, β C represents the random variable of urban water consumption forecast error.

3. The method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision-making according to claim 1 is characterized in that: The uncertainties include: The urban annual water consumption prediction model is used to predict the monthly water consumption of urban water plants within a year, as shown in the following formula: y pro (t)=g pro (t)+s pro (t)+h pro (t)+ε pro (t) Where y pro (t) represents the time series observation value at time t, g pro (t) represents the long-term trend term, s pro (t) represents the periodic term, h pro (t) represents the impact of holiday effects or other special events, ε pro (t) represents the error term; The comprehensive evaluation of the water quality of the external water source is to calculate the probability corresponding to the water quality level in each month, as shown in the following formula: In the formula, by lev，ind Indicates water quality type, lev indicates water quality standard level, ind indicates water quality index, bx ind Indicates the water quality index value, P Bayes (by lev，ind ) represents the prior probability. In the absence of monthly water quality information, the probabilities of different water quality types are equal. P Bayes (bx ind |by lev，ind ) represents the water quality index value bx ind Belong to the water quality type of this indicator lev，ind The possibility of P Bayes (by lev，ind |bx ind ) represents the posterior probability, indicating that the index value bx of the representative month is known ind Under certain conditions, the probability that the water quality of this indicator belongs to the lev level; The raw water diversion source quantity analysis model is used to determine the amount of water that can be diverted externally at different water inflow frequencies and the amount of water that can be diverted externally over many years based on existing relevant program results and historical measured data; The external water source and water use scenario generation model is used for the joint scheduling of multiple water sources in the city. The model includes a first-level urban external water source and regulating reservoir joint scheduling model, which rationally allocates the water inflow and outflow of cross-basin external water sources and regulating reservoirs; and a second-level water receiving unit optimization scheduling model, which allocates the water inflow and outflow of cross-basin external water sources and regulating reservoirs to different raw water plants for redistribution.

4. The method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision-making according to claim 1 is characterized in that: The method also includes an analysis model for the amount of raw water that can be diverted externally, which takes the results of existing relevant plans and historical measured data as input to determine the amount of water that can be diverted externally at different water inflow frequencies and the amount of water that can be diverted externally over many years as output.

5. According to the method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision-making according to claim 1, the constraints of the joint scheduling model of the water supply city regulating reservoir and external water source in step 2 include: i. The water balance constraint of external water source is expressed as follows: Where, is the total amount of externally diverted water from type m of externally diverted water sources in period t, is the water consumption of the raw water plant from the mth type of external water source in period t, is the reservoir inflow of type m external water source in period t, is the ecological water replenishment of the river from the m-type external water source in period t; ii. The water balance constraint of the regulating reservoir is expressed as follows: Where, is the water storage capacity of the rth reservoir at the end of the t-1 period, is the water storage capacity of the rth reservoir at the end of the tth period, is the inflow of the rth reservoir in the tth period, is the water supply from the rth reservoir to the raw water plant in the tth period, The leakage of the rth reservoir in the i-th period, Evaporation of the rth reservoir in period i; iii. Water transmission loss constraint, expressed as follows: Where, and The water loss coefficient for water transfer from regulating reservoirs and external water sources to urban water users; iv. Water supply guarantee rate constraint, expressed as follows: P urca1 ≤P urob ≤P urca2 Where, P urob is the calculated urban water supply guarantee rate, P urca1 and P urca2 To ensure the lower and upper limits of the rate for the set urban water supply targets; v. Reservoir capacity constraint, expressed as follows: Where, is the lower limit of the storage capacity of the rth reservoir in the tth period. The lower limit of the storage capacity of the reservoir is the dead storage capacity or the reduced storage capacity. is the upper limit of the water storage capacity of the rth reservoir in the tth period. The upper limit of the water storage capacity of the reservoir is the storage capacity corresponding to the flood limit water level during the flood season and the storage capacity corresponding to the normal water level during the non-flood season; vi. Pipeline water delivery capacity constraint, expressed as follows: WPE n,t ≤DE n Where WPE n，t is the water delivery of the nth water pipeline in the tth period, DE n is the actual maximum water delivery capacity of the nth water pipeline; vii. Raw water quality constraints: When the water quality index does not meet the surface water Class III standard and raw water emergency pretreatment is not considered, the amount of raw water that can be diverted is set to 0; viii. Non-negative constraints on variables: The model must satisfy the non-negative constraints on decision variables.

6. The method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision making according to claim 1 is characterized in that: The objective function for optimizing scheduling within the water receiving unit in step 2 is expressed as follows: Where: is the comprehensive relative water shortage of the raw water plant, DM p，t is the water demand of the pth water plant during period t; WTI j，p，t is the water supply from the jth water source to the pth water plant in the tth period, T is the period number, P is the raw water plant number, α p is the importance parameter of the pth water plant; The optimal scheduling constraint function within the water receiving unit is defined as follows: i. Constraints on available water supply from water sources, expressed as follows: The total amount of water supplied by the source to each water user shall not exceed the water supply capacity of the source. The water supply capacity constraint refers to the water diverted from the Yangtze River, the Luan River, and the storage capacity of the regulating reservoir: WTI j,p,t ≤SC j,p Where: SC j，p The maximum water supply capacity of the jth type of water source to the pth water plant; ii. The water supply capacity constraint of the pipeline project is expressed as follows: PS k,t ≤GC k,t Where, PS k，t is the water delivery of the kth water delivery channel in the tth period, GC k，t is the actual maximum water delivery capacity of the kth water pipeline; iii. Constraints on the water purification capacity of the water plant during each period are expressed as follows: DM p,t ≤PC p,t Where: PC p，t is the maximum water purification capacity of the water plant p at time t; iv. Raw water quality constraint, expressed as follows: When the water quality index does not meet the surface water Class III standard and the raw water emergency pretreatment is not considered, the raw water source that can be transferred externally is set to 0; v. The water supply constraint is a decision variable that satisfies the non-negative constraint. The expression is as follows: WTI j,p,t >0。 7. The method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision making according to claim 1 is characterized in that: The method further includes generating a scheduling scenario set of multiple uncertainties of the urban multi-source water supply by a Latin hypercube sampling method; the steps are as follows: For N uncertainty factor variables corresponding to N dimensions, each uncertainty factor variable is divided into M intervals with equal probability; a random uncertainty factor variable value is selected in each interval of each dimension, ensuring that each interval has one and only one sample value; the above steps are repeated until the required number of sample points are generated to cover the entire uncertainty factor parameter space.

8. The method for joint scheduling of multiple urban water sources based on discrete scenario sets and robust decision-making according to claim 1 is characterized in that: The joint scheduling model for urban regulating and storage reservoirs and external water sources within the year includes two objective functions: minimizing the risk of urban water shortage and minimizing the risk of water level drawdown in the regulating and storage reservoir at the end of the year. Under constraints, it outputs the allocation of external water transfer volume and the inflow and outflow of the regulating and storage reservoir. Based on the available water volume determined by the joint scheduling model for urban regulating and storage reservoirs and external water sources, and with the water demand of each user in the water receiving unit and the water transmission capacity of the pipeline as constraints, the water transmission volume of the raw water pipeline in each time period in the future is determined, thereby realizing the allocation of various water sources among different users in each water receiving unit.

Images