Multi-water-source joint scheduling analysis method based on water storage balance

Abstract

A multi-water-source joint scheduling analysis method based on water storage balance comprises the following steps: S10, acquiring data such as incoming water forecast, water demand prediction results, water source real-time water levels and reservoir capacities, and performing real-time daily adjustment calculation; s20, improving a particle swarm optimization algorithm through a multi-water-source water storage rate equalization algorithm, and generating a non-flood season water source water storage equalization optimal scheduling scheme; s30, improving a particle swarm algorithm through a multi-water-source water storage rate equalization algorithm, and generating a flood season abandoned water resource utilization optimal scheduling scheme; and S40, through a multi-water-source water storage rate equalization algorithm, improving a particle swarm algorithm, and generating a flood season abandoned water resource utilization and flood prevention regulation control bidirectional optimization optimal scheduling scheme. The method is suitable for multi-water-source joint scheduling analysis calculation through various optimization scheduling target scenes, can support real-time analysis of the step length of any day time period, further excavates the water resource quantity, and guarantees the water supply safety of a water receiving area and the safe operation of a water network project.

Description

Technical Field

[0001] The present invention belongs to the technical field of multi-water source scheduling analysis, and in particular relates to a multi-water source joint scheduling analysis method based on water storage balance. Background Art

[0002] The joint dispatch analysis of multiple water sources is an important requirement for application scenarios of real-time dispatch, planning and optimization of water resources. The existing joint dispatch analysis methods of multiple water sources mainly include water inflow forecast module, water demand forecast module and dynamic dispatch model. Based on the topological structure of the water supply network of "water source-water plant-water user", it can realize the coordinated water supply and demand analysis of different scales such as regional year and month, and complete the dynamic dispatch of three levels of region, water source and project on this basis.

[0003] At the same time, the current multi-water source joint scheduling analysis method still has some defects. For example, the optimization and allocation target mainly considers minimizing the water shortage rate at the user end, which makes it difficult to ensure the water supply safety at the water source end; the calculation time step is mostly years and months, which makes it difficult to ensure real-time scheduling operations; the combination of water resource utilization and flood prevention during the flood season is not considered, and the ability to tap water resources is insufficient.

[0004] Optimized multi-water source joint scheduling analysis can promote the optimal utilization and precise scheduling of regional water resources, support real-time analysis of any daily time period and step length, further tap the amount of water resources, better meet the actual application needs of water resource allocation business in various places, and ensure the water supply safety of the receiving areas and the safe operation of water network projects.

[0005] Based on the above situation, this application proposes a multi-water source joint scheduling analysis method based on water storage balance. Summary of the invention

[0006] The present invention provides a multi-water source joint scheduling analysis method based on water storage balance, which is suitable for multi-water source joint scheduling analysis and calculation through a variety of optimization scheduling target scenarios, can support real-time analysis of any daily time period step, further explore the amount of water resources, better meet the actual application needs of water resource allocation business in various places, and ensure the water supply safety of the receiving area and the safe operation of the water network project.

[0007] The present invention is solved by the following technical solutions.

[0008] A multi-water source joint scheduling analysis method based on water storage balance includes the following steps: S10: obtaining water inflow forecast, water demand forecast results, water source real-time water level and storage capacity data, and performing real-time daily adjustment calculation; executing at least one of the following steps S20, S30, and S40: S20: using a multi-water source water storage rate balance algorithm and an improved particle swarm algorithm to achieve water storage balance in non-flood seasons, performing self-learning iterative optimization, and generating an optimal scheduling plan for water storage balance in non-flood seasons; S30: using a multi-water source water storage rate balance algorithm and an improved particle swarm algorithm to achieve water storage balance in non-flood seasons. With the goal of realizing the utilization of abandoned water resources during the flood season, further self-learning iterative optimization is performed through the water storage resource optimization utilization model, and the optimal scheduling plan for the utilization of abandoned water resources during the flood season is generated; S40: Through the multi-water source water storage rate balancing algorithm and the improved particle swarm algorithm, with the goal of realizing the two-way optimization of the utilization of abandoned water resources during the flood season and flood control and regulation control, further self-learning iterative optimization is performed through the water storage resource optimization utilization model, the water source flood and rainfall capacity dynamic analysis model, and the two-way optimization and allocation coupling model, and the optimal scheduling plan for the two-way optimization of the utilization of abandoned water resources during the flood season and flood control and regulation control is generated.

[0009] In a preferred embodiment, the multi-water source water storage rate balancing algorithm: the water storage rates of each water source are relatively balanced at the end of the calculation period, ensuring that the relative dispersion of the water storage rates of each water source of interest is minimized. Specifically, the objective function of the multi-water source water storage rate balancing algorithm is as follows:

[0010] ZY=Min ;

[0011] in : Impact factor; : Water source storage rate; : Average water storage rate of n water sources; n: number of water sources.

[0012] Furthermore, in the objective function of the multi-water source water storage rate balancing algorithm:

[0013]

[0014]

[0015]

[0016]

[0017] in: for Final storage capacity of water source during the period; for The final storage capacity of the time period is also equivalent to Initial storage capacity of the time period; for The amount of water entering the reservoir from the water source during a period can be determined according to the formula; is the amount of evaporated water from the water source; is the infiltration volume of water source; For the period The water supply from the source; The amount of water that is discarded exceeds the water storage capacity of the water source. Normal storage capacity; is the water supply from a water source to user j, water use department b, in the kth period, js is the number of users, and bs is the number of water use departments; To improve the position of iterative particles in the particle swarm algorithm; the units of the above storage capacity, water volume, and infiltration volume are all m 3 .

[0018] In a preferred embodiment, the improved particle swarm algorithm generates multiple populations of high-dimensional particles for the water supply from the water source to the water plant in each control period during the water supply scheduling period, combines the constructed simulation water resource allocation model to perform supply and demand balance calculations, analyzes the discrete degree of the water storage rate of each water source in each period during the scheduling period, continuously iterates the particles to learn and optimize, and feedback is used to correct the water resource allocation model parameters to obtain an optimized water supply allocation plan from the water source to the water plant.

[0019] In a preferred implementation, the improved particle swarm algorithm includes a speed update algorithm and a position update algorithm; the formula of the speed update algorithm is as follows:

[0020]

[0021] in:

[0022] : The velocity of the ith particle in the t+1th iteration in the dth dimension;

[0023] ω: Inertia weight, used to balance the particle's dependence on its previous velocity;

[0024] : The velocity of the ith particle in the tth iteration in the dth dimension;

[0025] c1 and c2: learning factors or acceleration constants, representing the weights of learning from the individual optimal solution and the global optimal solution respectively;

[0026] r1 and r2: random numbers between [0,1];

[0027] : The individual historical optimal position of the i-th particle in the d-th dimension;

[0028] : The global optimal position of all particles in the dth dimension;

[0029] : The position of the ith particle at the tth iteration in the dth dimension;

[0030] The formula of the location update algorithm is as follows:

[0031]

[0032] in: : The position of the ith particle at the t+1th iteration in the dth dimension.

[0033] In the improved particle swarm algorithm, each particle represents a potential solution to the problem, that is, the water supply from the water source to the water plant, and searches for the optimal solution by iteratively updating its speed and position. The three parts in the speed update formula represent the inheritance of the previous speed (inertia), self-cognition (the influence of the individual optimal solution), and social cognition (the influence of the global optimal solution). This update mechanism enables the particle swarm to effectively explore and develop in the search space to find the optimal solution to the problem.

[0034] In a preferred embodiment, the water storage resource optimization utilization model is based on the actual water supply regulation and control operation of urban water sources and water plants. For the networked water supply of reservoir groups, the global and single hierarchical interactive parallel optimization calculation methods are used to decouple and analyze the water supply optimization goals of multiple water sources to multiple water plant users under multiple rule restrictions, ensure that the single water source meets the conditions for abandoned water utilization during each water supply regulation period, and increase water supply, ensure that the entire water source group meets the relative balance of water storage during each water supply regulation period and the relative maximization of the total water storage at the end of the scheduling period, so as to obtain an optimized water resource utilization plan.

[0035] In a preferred embodiment, the dynamic analysis model of the water source's flood and rainfall capacity takes into account the actual flood control scheduling and operation of regional water sources, sets the upper limit control water level for each water source to receive floods and rainfall (such as flood control high water level, design flood level or other set water level, etc.), or sets a flood warning control rainfall value, and combines the discharge capacity of each water source's discharge facilities and spillways and the hydrological characteristics of each water source basin to calculate the rainfall capacity of each water source at the water level and judge the degree to which the warning control rainfall is met.

[0036] In a preferred embodiment, the dynamic analysis model of the water source's flood and rain-bearing capacity includes a rainfall-runoff empirical correlation graph model, a water source runoff depth calculation model, and a carryable rainfall calculation model.

[0037] Compared with the prior art, the present invention has the following beneficial effects: it provides a multi-water source joint scheduling analysis method based on water storage balance, which is suitable for multi-water source joint scheduling analysis and calculation through a variety of optimization scheduling target scenarios, and can support real-time analysis of any daily time period step, further tap the amount of water resources, better meet the actual application needs of water resource allocation business in various places, and ensure the water supply safety of the receiving area and the safe operation of the water network project. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of the technical solution in the present invention.

[0039] Figure 2 It is the three-variable correlation diagram of rainfall-runoff experience in the present invention. DETAILED DESCRIPTION

[0040] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0041] In the following embodiments, the same or similar reference numerals throughout represent the same or similar components or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0042] In the description of the present invention, it should be understood that the terms: center, longitudinal, transverse, length, width, thickness, up, down, front, back, left, right, vertical, horizontal, top, bottom, inside, outside, clockwise, counterclockwise, etc. indicating the orientation or position relationship are based on the orientation or position relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, and therefore cannot be understood as limiting the present invention. In addition, the terms: first, second, etc. are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features shown. In the description of the present invention, unless otherwise clearly specified and limited, the terms: install, connect, connect, etc. should be understood in a broad sense, and ordinary technicians in this field can understand the specific meanings of the above terms in this application according to the specific circumstances.

[0043] See also Figure 1 to Figure 2A multi-water source joint scheduling analysis method based on water storage balance includes the following steps: S10: obtaining water inflow forecast, water demand forecast results, water source real-time water level and storage capacity data, and performing real-time daily adjustment calculation; executing at least one of the following steps S20, S30, and S40: S20: using a multi-water source water storage rate balance algorithm and an improved particle swarm algorithm to achieve water storage balance in non-flood seasons, performing self-learning iterative optimization, and generating an optimal scheduling plan for water storage balance in non-flood seasons; S30: using a multi-water source water storage rate balance algorithm and an improved particle swarm algorithm, With the goal of realizing the utilization of abandoned water resources during the flood season, self-learning iterative optimization is further performed through the water storage resource optimization utilization model, and the optimal scheduling plan for the utilization of abandoned water resources during the flood season is generated; S40: Through the multi-water source water storage rate balancing algorithm and the improved particle swarm algorithm, with the goal of realizing the two-way optimization of the utilization of abandoned water resources during the flood season and flood control and regulation control, self-learning iterative optimization is further performed through the water storage resource optimization utilization model, the water source flood and rainfall capacity dynamic analysis model, and the two-way optimization and allocation coupling model, and the optimal scheduling plan for the two-way optimization of the utilization of abandoned water resources during the flood season and flood control and regulation control is generated.

[0044] Specifically, in this application, the multi-water source water storage rate balancing algorithm: the water storage rates of each water source are relatively balanced at the end of the calculation period, ensuring that the relative discreteness of the water storage rates of each water source reservoir capacity of concern is minimized. The objective function of the multi-water source water storage rate balancing algorithm is as follows:

[0045] ZY=Min ;

[0046] in : Impact factor; : Water source storage rate; : Average water storage rate of n water sources; n: number of water sources.

[0047] Furthermore, in the objective function of the multi-water source water storage rate balancing algorithm:

[0048]

[0049]

[0050]

[0051]

[0052] in: for Final storage capacity of water source during the period; for The final storage capacity of the time period is also equivalent to Initial storage capacity of the time period; for The amount of water entering the reservoir from the water source during a period can be determined according to the formula; is the amount of evaporated water from the water source; is the infiltration volume of water source; For the period The water supply from the source; The amount of water that is discarded exceeds the water storage capacity of the water source. Normal storage capacity; is the water supply from a water source to user j, water use department b, in the kth period, js is the number of users, and bs is the number of water use departments; To improve the position of iterative particles in the particle swarm algorithm; the units of the above storage capacity, water volume, and infiltration volume are all m 3 .

[0053] In the present application, the improved particle swarm algorithm generates multiple populations of high-dimensional particles for the water supply from the water source to the water plant in each control period during the water supply scheduling period, combines the constructed simulation water resource allocation model to perform supply and demand balance calculations, analyzes the discrete degree of the water storage rate of each water source in each period during the scheduling period, and continuously iterates the particles to learn and optimize, and feedback is used to correct the water resource allocation model parameters to obtain an optimized water supply allocation plan from the water source to the water plant.

[0054] Specifically, the improved particle swarm algorithm includes a speed update algorithm and a position update algorithm; the formula of the speed update algorithm is as follows:

[0055]

[0056] in:

[0057] : The velocity of the ith particle in the t+1th iteration in the dth dimension;

[0058] ω: Inertia weight, used to balance the particle's dependence on its previous velocity;

[0059] : The velocity of the ith particle in the tth iteration in the dth dimension;

[0060] c1 and c2: learning factors or acceleration constants, representing the weights of learning from the individual optimal solution and the global optimal solution respectively;

[0061] r1 and r2: random numbers between [0,1];

[0062] : The individual historical optimal position of the i-th particle in the d-th dimension;

[0063] : The global optimal position of all particles in the dth dimension;

[0064] : The position of the ith particle at the tth iteration in the dth dimension;

[0065] The formula of the location update algorithm is as follows:

[0066]

[0067] in: : The position of the ith particle at the t+1th iteration in the dth dimension.

[0068] In the improved particle swarm algorithm, each particle represents a potential solution to the problem, that is, the water supply from the water source to the water plant, and searches for the optimal solution by iteratively updating its speed and position. The three parts in the speed update formula represent the inheritance of the previous speed (inertia), self-cognition (the influence of the individual optimal solution), and social cognition (the influence of the global optimal solution). This update mechanism enables the particle swarm to effectively explore and develop in the search space to find the optimal solution to the problem.

[0069] In addition, in the present application, the water storage resource optimization utilization model is based on the actual water supply regulation and control operation of urban water sources and water plants. The model meets the settings of any total scheduling period (such as 30 days, 45 days, etc.) and actual water supply regulation calculation period (such as 1 day, 3 days, 5 days, etc.). For the networked water supply of reservoir groups, the global and single hierarchical interactive parallel optimization calculation methods are used to decouple and analyze the water supply optimization goals of multiple water sources to multiple water plant users under multiple rule restrictions. If the single water source meets the conditions for abandoned water utilization during each water supply regulation period, the water supply will be increased, and the entire water source group will meet the relative balance of water storage during each water supply regulation period and the relative maximization of the total water storage at the end of the scheduling period, thereby obtaining an optimized water resource utilization plan.

[0070] The water storage resource optimization utilization model is used to achieve the following two goals:

[0071]

[0072]

[0073] in: for Final water storage capacity of the period (m 3 ); : Impact factor; : Water source storage rate; :The average water storage rate of n water sources n is the number of water sources; Q1 abandoned water utilization - increase water supply: reservoir water level Z> abandoned water setting water level Z m (Such as flood limit water level or other set water level).

[0074] In addition, in the present application, the dynamic analysis model of the water source's flood and rainfall capacity takes into account the actual flood control scheduling and operation of the regional water source, sets the upper limit control water level for each water source to receive floods and rainfall (such as flood control high water level, design flood level or other set water level, etc.), or sets a flood warning control rainfall value, and combines the discharge capacity of each water source's discharge facilities and spillways and the hydrological characteristics of each water source basin to calculate the rainfall capacity of each water source at the water level and judge the degree to which the warning control rainfall is met.

[0075] Furthermore, the dynamic analysis model of the water source's flood and rain-bearing capacity includes a rainfall-runoff empirical correlation diagram model, a water source runoff depth calculation model and a carrying rainfall calculation model, as described below.

[0076] (1) Empirical correlation graph model of rainfall runoff

[0077] The model is based on the physical mechanism of rainfall and runoff in the basin, takes the main influencing factors as parameters, and establishes a quantitative correlation between rainfall P and runoff R. Commonly used parameters include the previous rainfall index P a (the soil is wet in the early stage of the reaction), etc. The formula is: , currently used is a three-variable correlation diagram, such as Figure 2 shown.

[0078] Figure 2 The characteristics are: a The curve cluster is on the upper left side of the 45° straight line, P a The larger it is, the closer it is to the 45° line, that is, the smaller the rainfall loss is; each P a There is a turning point for all contour lines. The relationship line above the turning point is a 45° straight line, and the relationship line below the turning point is a curve. a The horizontal spacing between straight line segments is equal.

[0079] Parameter P a Generally, it is calculated using the empirical formula: In the formula, is the previous rainfall index at 8 a.m. on t day; n is the number of previous rainfall days that affect the runoff, usually around 15 days; k is a constant coefficient, which can generally be around 0.85.

[0080] (2) Calculation model of water source runoff depth

[0081] The calculation of reservoir runoff depth is mainly divided into two parts. One part is the runoff depth converted from the free storage capacity when the reservoir does not discharge flood water, and the other part is the runoff depth converted from the flood discharge volume when the reservoir discharges flood water, that is: In the formula, R T is the total runoff depth, R L is the free storage capacity equivalent to the runoff depth, R f It is the flood discharge converted to runoff depth.

[0082] Specifically: Free storage capacity converted to runoff depth: Assuming that the reservoir does not discharge flood water, calculate the free storage capacity W between the starting water level (i.e. the water level at the forecast time) and the specific water level L According to the reservoir engineering characteristic data, the starting storage capacity W can be calculated according to the starting water level. S , and then use the specific water level to correspond to the reservoir capacity W T Subtract the initial storage capacity W S , that is, the available storage capacity W L， .

[0083] The conversion relationship between basin runoff and runoff depth is: Where F is the reservoir catchment area, unit: km 2 .

[0084] Calculate the free storage capacity equivalent to the runoff depth R L , .

[0085] Specifically: Free storage capacity converted to runoff depth: The discharge of a reservoir mainly includes two parts, one is the discharge from the water diversion tunnel and the other is the discharge from the spillway. Based on the engineering characteristics of each reservoir in the basin, the characteristic curve of the reservoir water level ~ maximum discharge of the discharge structure of each reservoir is obtained. The reservoir calculates the discharge W from the starting water level to a specific water level through the relationship between the reservoir water level and the maximum discharge of the discharge structure or the measured discharge. f , determined according to the reservoir discharge form, the discharge volume is calculated to be equivalent to the runoff depth, .

[0086] Furthermore, the two-way optimization coupling model in the present application is as follows: by coupling the water storage resources optimization utilization model and the water source flood and rainfall capacity dynamic analysis model, for each control period within the scheduling water supply period, the flood and rainfall capacity satisfaction degree of each water source and the water abandonment resource situation are automatically calculated and analyzed, and the corresponding water supply weights are continuously optimized and increased, so as to obtain a raw water allocation plan with relatively better comprehensive goals.

[0087] Goal 1 ;

[0088]

[0089] for End storage capacity of the reservoir during the period (m 3 ); : Impact factor; : Reservoir water storage rate; n is the number of water sources; Q1 abandoned water utilization - increase water supply: reservoir water level Z> abandoned water setting water level Z m(such as flood limit water level or other set water level); Q2 Flood control - increase water supply: the reservoir can accommodate rainfall P < the reservoir flood control rainfall requirement P m .

[0090] The technical solution in this application obtains data such as water inflow forecast, water demand forecast results, real-time water level and storage capacity of water sources from the original water resources dynamic allocation model, considers the hydrological characteristics of each water source in the region, and constructs a multi-water source water storage rate balance algorithm based on the improved particle swarm algorithm. Based on the actual scheduling application needs, through real-time daily adjustment calculation, it realizes the joint scheduling analysis of multiple water sources with any daily time step (1 day, 3 days, 5 days, etc.). With the goal of balancing water storage of non-flood season water sources, utilizing abandoned water resources in flood season, and bidirectional optimization of abandoned water resources utilization and flood control regulation control in flood season, it integrates the water storage resource optimization utilization model, the water source flood and rain capacity dynamic analysis model and the bidirectional optimization allocation coupling model, performs self-learning iterative optimization calculation, and generates the corresponding optimal multi-water source joint scheduling plan.

[0091] The technical solution in this application has the following beneficial effects.

[0092] (1) In the non-flood season, based on the premise of ensuring regional water demand, the hydrological characteristics of each water source in the region are considered, and a multi-source water storage rate equilibrium algorithm based on improved particle swarm is constructed to achieve joint optimization scheduling of multiple water sources and improve the ability to ensure regional water supply security;

[0093] (2) During the flood season, based on the multi-source water storage rate balance algorithm based on the improved particle swarm, the water storage resource optimization utilization model is integrated to further optimize the joint scheduling of multiple water sources with the goal of tapping the abandoned water resources during the flood season;

[0094] (3) Aiming at the two-way problems of flood control dispatch and water resources allocation faced by my country's coastal areas during the typhoon season, based on the multi-source water storage rate balance algorithm based on the improved particle swarm, the water storage resource optimization utilization model, the water source flood and rain capacity dynamic analysis model and the two-way optimization allocation coupling model are integrated to achieve the two-way optimization allocation of abandoned water resource utilization and flood control regulation control with the goal of ensuring regional flood control safety and efficient use of water resources;

[0095] (4) Based on actual scheduling application needs, real-time daily adjustment calculation is used to realize multi-water source joint scheduling analysis with any daily time step (1 day, 3 days, 5 days, etc.).

[0096] The protection scope of the present invention includes but is not limited to the above embodiments. The protection scope of the present invention shall be based on the claims. Any replacement, deformation, and improvement of the technology that can be easily thought of by technicians in this field shall fall within the protection scope of the present invention.

Claims

1. A multi-water source joint scheduling analysis method based on water storage balance, characterized in that: The following steps are involved: S10: Obtain water inflow forecast, water demand forecast results, real-time water level and reservoir capacity of water sources and other data to perform real-time daily regulation calculations; Execute at least one of the following steps S20, S30, and S40: S20: Through the multi-water source water storage rate balance algorithm and the improved particle swarm algorithm, the self-learning iterative optimization is carried out to achieve the water storage balance of the non-flood season water source, and the optimal scheduling plan for the water storage balance of the non-flood season water source is generated; S30: Through the multi-source water storage rate balance algorithm and the improved particle swarm algorithm, the goal is to achieve the utilization of abandoned water resources during the flood season, and further self-learning iterative optimization is performed through the water storage resource optimization utilization model to generate the optimal scheduling plan for the utilization of abandoned water resources during the flood season; S40: Through the multi-water source water storage rate balancing algorithm and the improved particle swarm algorithm, the goal is to achieve two-way optimization of the utilization of abandoned water resources during the flood season and flood control and regulation control. Further self-learning iterative optimization is carried out through the water storage resource optimization utilization model, the water source flood and rainfall capacity dynamic analysis model, and the two-way optimization and allocation coupling model, and the optimal scheduling plan for the two-way optimization of the utilization of abandoned water resources during the flood season and flood control and regulation control is generated.

2. The multi-water source joint scheduling analysis method based on water storage balance according to claim 1 is characterized in that: The multi-water source storage rate balancing algorithm: the water storage rates of various water sources are relatively balanced at the end of the calculation period, ensuring that the relative dispersion of the water storage rates of various water source reservoirs of concern is minimized.

3. The multi-water source joint scheduling analysis method based on water storage balance according to claim 2 is characterized in that: The objective function of the multi-source water storage rate balancing algorithm is as follows: ZY=Min ; in : Impact factor; : Water source storage rate; : Average water storage rate of n water sources; n: number of water sources.

4. The multi-water source joint scheduling analysis method based on water storage balance according to claim 3 is characterized in that: In the objective function of the multi-water source water storage rate balancing algorithm: in: for Final storage capacity of water source during the period; for The final storage capacity of the time period is also equivalent to Initial storage capacity of the time period; for The amount of water entering the reservoir from the water source during a period can be determined according to the formula; is the amount of evaporated water from the water source; is the infiltration volume of water source; For the period The water supply from the source; The amount of water that is discarded exceeds the water storage capacity of the water source. Normal storage capacity; is the water supply from a water source to user j, water use department b, in the kth period, js is the number of users, and bs is the number of water use departments; To improve the position of iterative particles in the particle swarm algorithm; The units of the above storage capacity, water volume and infiltration volume are all m 3 .

5. The multi-water source joint scheduling analysis method based on water storage balance according to claim 1 is characterized in that: The improved particle swarm algorithm generates multiple populations of high-dimensional particles for the water supply from the water source to the water plant in each control period during the water supply scheduling period, combines the constructed simulation water resource allocation model to perform supply and demand balance calculations, analyzes the discrete degree of the water storage rate of each water source in each period during the scheduling period, continuously iterates particles to learn and optimize, and feedbacks and corrects the parameters of the water resource allocation model to obtain an optimized water supply allocation plan from the water source to the water plant.

6. The multi-water source joint scheduling analysis method based on water storage balance according to claim 5 is characterized in that: The improved particle swarm algorithm includes a speed update algorithm and a position update algorithm; The formula of the speed update algorithm is as follows: in: : The velocity of the ith particle in the t+1th iteration in the dth dimension; ω: Inertia weight, used to balance the particle's dependence on its previous velocity; : The velocity of the ith particle in the tth iteration in the dth dimension; c1 and c2: learning factors or acceleration constants, representing the weights of learning from the individual optimal solution and the global optimal solution respectively; r1 and r2: random numbers between [0,1]; : The individual historical optimal position of the i-th particle in the d-th dimension; : The global optimal position of all particles in the dth dimension; : The position of the ith particle at the tth iteration in the dth dimension; The formula of the location update algorithm is as follows: in: : The position of the ith particle at the t+1th iteration in the dth dimension.

7. The multi-water source joint scheduling analysis method based on water storage balance according to claim 1 is characterized in that: The water storage resource optimization utilization model is based on the actual water supply regulation and operation of urban water sources and water plants. It uses global and single hierarchical interactive parallel optimization calculation methods for networked water supply of reservoir groups, decouples and analyzes the water supply optimization goals of multiple water sources to multiple water plant users under multiple rule restrictions, ensures that the single water source meets the abandoned water utilization conditions during each water supply regulation period, and increases water supply, ensures that the entire water source group meets the relative balance of water storage during each water supply regulation period and the relative maximization of the total water storage at the end of the scheduling period, so as to obtain an optimized water resource utilization plan.

8. The multi-water source joint scheduling analysis method based on water storage balance according to claim 1 is characterized in that: The dynamic analysis model of water source flood and rainfall capacity takes into account the actual flood control scheduling and operation of regional water sources, sets the upper limit control water level of each water source, or sets the flood warning control rainfall value, and combines the discharge capacity of each water source's discharge facilities and spillways as well as the hydrological characteristics of each water source basin to calculate the rainfall capacity of each water source at the water level and judge the degree to which the warning control rainfall is met.

9. The multi-water source joint scheduling analysis method based on water storage balance according to claim 8 is characterized in that: The dynamic analysis model of the water source's flood and rain-bearing capacity includes a rainfall-runoff empirical correlation diagram model, a water source runoff depth calculation model and a carryable rainfall calculation model.

Images