A Variable-Scale Chaotic Search Optimization Method and System for Drawdown Scheduling of Cascade Reservoir Groups

Abstract

This invention relates to the field of efficient water resource utilization and optimized hydropower scheduling technology, and discloses a variable-scale chaotic search optimization method for drawdown scheduling of cascade reservoir groups. Given an initial drawdown scheduling process for the reservoir group, the method decomposes the multi-stage drawdown scheduling optimization problem into several two-stage progressive optimization sub-problems. For each progressive optimization sub-problem, a chaotic search strategy of coarse search or fine search is selected with a certain probability for optimization. The method iteratively calculates and optimizes to successively approach the global optimum, and outputs the optimal drawdown scheduling process. This invention also discloses a variable-scale chaotic search optimization system for drawdown scheduling of cascade reservoir groups. This invention's variable-scale chaotic search optimization method and system for drawdown scheduling of cascade reservoir groups applies chaotic search to the two-stage optimization problem of the Power of Attention (POA), and fully extracts and captures solution space information through variable-scale chaotic mapping, further improving the solution speed and computational accuracy of the POA.

Description

Technical Field

[0001] This invention relates to the field of efficient water resource utilization and optimized hydropower scheduling technology, specifically to a variable-scale chaotic search optimization method and system for drawdown scheduling of cascade reservoir groups. Background Technology

[0002] my country's water conservancy industry has developed rapidly, with a number of large-capacity reservoirs such as Wudongde and Baihetan reservoirs being built and put into operation. These reservoirs undertake comprehensive utilization tasks including flood control, power generation, navigation, ecological restoration, and dry season water replenishment, which are of great significance to promoting social and economic development and implementing the national energy strategy. However, with the construction and operation of large-scale reservoir groups, the dry season runoff of the basin has been significantly affected. The reservoir operation modes arranged in the preliminary design stage are no longer fully adapted to the new situation. It is necessary to focus on the dry season operation characteristics and drawdown modes of large reservoir groups, and rationally arrange the drawdown sequence, timing, and process of reservoir groups to fully realize the comprehensive utilization benefits of water resources. Therefore, the construction and efficient solution of the drawdown optimization scheduling model for cascade reservoir groups is an important part of the joint scheduling of reservoir groups and a key link between the dry season and the flood season, which has received widespread attention from operation and management departments.

[0003] Given the runoff process during the drawdown period, the optimal drawdown scheduling of a reservoir group aims to meet multiple scheduling needs, including water supply, ecology, and navigation, and maximize the power generation of the cascade reservoir group. However, the optimal drawdown scheduling of a cascade reservoir group is a typical complex, high-dimensional, nonlinear, and strongly constrained optimization problem, which is difficult to solve. In addition, the increasing number of reservoirs leads to a significant increase in the dimensionality of the joint optimization scheduling problem, and traditional dynamic programming methods suffer from the curse of dimensionality, making them difficult to apply effectively.

[0004] The Progressive Optimization Approach (POA) is an improved dynamic programming method for solving multi-stage, multi-state optimization problems. Its basic idea is to decompose the multi-stage decision optimization problem into a series of two-stage optimization problems. In each computational stage, the variables of other stages are fixed, and the objective function value is optimized only for the current stage and the two adjacent stages. This process is repeated iteratively until the termination condition is met and the final result is output. This method reduces the discrete states of feasible solutions and significantly improves computational efficiency compared to dynamic programming. However, when there are many reservoirs, it can fall into the curse of dimensionality, exhibiting drawbacks such as high computational time overhead, limited global search range, and insufficient granularity of local searches. Therefore, there is an urgent need to effectively improve POA to enhance the optimization accuracy and computational efficiency of the joint scheduling problem of reservoir groups.

[0005] Chaos is a common phenomenon in nonlinear systems, characterized by regularity, randomness, and ergodicity. In recent years, various chaos search strategies have emerged. Their basic idea is to transform the chaotic space and the solution space, and to use the regularity, randomness, and ergodicity of chaotic variables to perform iterative searches. These strategies have advantages such as fast convergence, global asymptotic convergence, and easy escape from local optima, making them very suitable for combination with other computational methods. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of the aforementioned technologies by providing a variable-scale chaotic search optimization method and system for drawdown scheduling of cascade reservoir groups. This method applies chaotic search to the two-stage optimization problem of Problem of Arithmetic Algorithms (POA), fully extracts and captures solution space information through variable-scale chaotic mapping, applies coarse-scale chaotic search in the early stage of iteration to enhance the diversity of the optimization process, and applies fine-scale chaotic search in the later stage of iteration to mine local effective information, thereby further improving the solution speed and computational accuracy of POA.

[0007] To achieve the above objectives, the present invention provides a variable-scale chaotic search optimization method for drawdown scheduling of cascade reservoir groups, comprising the following steps:

[0008] 1) Determine the initial calculation conditions, select the reservoir group to participate in the joint drawdown scheduling, and clarify the objective function, constraints, and decision variables of the reservoir group drawdown optimization scheduling;

[0009] 2) Set the calculation parameters, including the maximum number of iterations M, the convergence accuracy ε, the total number of coarse-scale chaotic searches CH1 and the total number of fine-scale chaotic searches CH2 in the variable-scale chaotic search;

[0010] 3) Generate the initial water level process for the drawdown scheduling of the reservoir group based on human experience or conventional methods. And calculate the initial distance step length Δ = (Δ) for each reservoir. i,j ) N×T ,in Z i,j Let Δ be the water level of reservoir i during time period j. i,j For the distance of reservoir i from the time period j, and Let N be the upper and lower limits of the water level of reservoir i in time period j, respectively; N be the number of reservoirs; T be the number of time periods for scheduling; and K be the initial number of discrete nodes.

[0011] 4) Initialize the number of iterations m = 1;

[0012] 5) Decompose the reservoir group drawdown scheduling optimization problem of the current iteration number into T-1 two-stage stepwise optimization subproblems;

[0013] 6) For T-1 two-stage stepwise optimization subproblems, a coarse-scale chaotic search or a fine-scale chaotic search is randomly selected according to probability for optimization: when p > m / M, a coarse-scale chaotic search is executed; otherwise, a fine-scale chaotic search is executed, where p is a random number between 0 and 1.

[0014] 7) Initialize the time period j = 1;

[0015] 8) The water levels of each reservoir during time period j-1 are: Water levels of each reservoir during time period j+1 If p > m / M, for Perform coarse-scale chaotic search optimization; if p ≤ m / M, for Perform fine-scale chaotic search to find the optimal solution.

[0016] 9) Let j = j + 1. If j ≤ T - 1, go to step 8); otherwise, go to step 10.

[0017] 10) Calculate the scheduling process between two consecutive iterations. and like Then shrink all reservoirs away from the walking distance, making Proceed to step 11), otherwise proceed to step 5.

[0018] 11) Let m = m + 1. If m > M, then go to step 12); otherwise, go to step 5.

[0019] 12) Terminate the calculation and output the optimal water level process for each reservoir.

[0020] Preferably, step 8) involves performing a coarse-scale chaotic search for optimization, including the following steps:

[0021] 8.1) Normalization generates chaotic variables: the water levels of each reservoir in time period j are normalized. Normalization is performed to generate initial chaotic variables. The mapping is to the interval [0,1], where λ is the chaos control parameter. When λ = 4, it is in a completely chaotic state.

[0022] 8.2) Iterative generation of coarse-scale chaotic search variables: Based on the chaotic mapping generation method, the chaotic variables of the first coarse-scale chaotic search are generated. Chaotic variables in the first coarse-scale chaos search (ch1)

[0023] 8.3) Coarse-scale chaotic search optimization calculation: including the following steps:

[0024] 8.3.1) Initialize the number of coarse-scale chaotic searches ch1 = 1;

[0025] 8.3.2) Calculation The calculation method is as follows: Compare and like Superior make

[0026] 8.3.3) Let ch1 = ch1 + 1; if ch1 ≤ CH1, go to step 8.3.2); otherwise, go to step 9.

[0027] Preferably, in step 8), performing fine-scale chaotic search optimization includes the following steps:

[0028] 8.4) Normalization generates chaotic variables: the water levels of each reservoir in time period j are normalized. Normalization is performed to generate initial chaotic variables. To map to the interval [0,1];

[0029] 8.5) Iterative Generation of Fine-Scale Chaotic Search Variables: Based on the chaotic mapping generation method, the chaotic variables of the first fine-scale chaotic search are... The chaotic variables in the second fine-scale chaotic search are:

[0030] 8.6) Fine-scale chaotic search optimization calculation: including the following steps:

[0031] 8.6.1) Initialize the number of coarse-scale chaotic searches ch2 = 1;

[0032] 8.6.2) will Inverse normalization to Interval, calculation The calculation method is as follows: In the formula, η is an adaptive parameter, which is a random number between 0 and 1. and like Superior make

[0033] 8.6.3) Let ch2 = ch2 + 1; if ch2 ≤ CH2, go to step 8.6.2); otherwise, go to step 9.

[0034] A variable-scale chaotic search optimization system for drawdown scheduling of a cascade reservoir group, used in the variable-scale chaotic search optimization method for drawdown scheduling of the cascade reservoir group, includes:

[0035] 1) Initialization module: Determine the objective function, constraints and decision variables of the drawdown optimization scheduling of the cascade reservoir group, and generate the initial drawdown scheduling process of the reservoir group that meets the constraints based on human experience or conventional methods under various constraints;

[0036] 2) Stepwise optimization decomposition module: The reservoir group drawdown optimization scheduling problem of the current iteration number is decomposed into multiple two-stage stepwise optimization sub-problems;

[0037] 3) Variable-scale chaotic search selection module: It judges each stepwise optimization subproblem and randomly selects coarse-scale chaotic search or fine-scale chaotic search for optimization solution according to probability;

[0038] 4) Coarse-scale chaotic search module: Performs coarse-scale chaotic search, including normalization to generate chaotic variables, iterative generation of coarse-scale chaotic search variables, coarse-scale chaotic search optimization calculation, and stopping condition judgment functions, in order to obtain better optimization results;

[0039] 5) Fine-scale chaotic search module: Performs fine-scale chaotic search, including normalization to generate chaotic variables, iterative generation of fine-scale chaotic search variables, fine-scale chaotic search optimization calculation and stopping condition judgment functions, so as to obtain better optimization results;

[0040] 6) Output module: Repeatedly execute the stepwise optimization decomposition module, the variable-scale chaotic search selection module, the coarse-scale chaotic search module, and then the fine-scale chaotic search module until the preset iteration stopping condition is met, obtain the optimal scheduling result, and output it as the final cascade reservoir group drawdown optimization scheduling result.

[0041] Compared with the prior art, the present invention has the following advantages:

[0042] 1. Based on the stepwise optimization method, this computational framework combines variable-scale chaotic search and two-stage stepwise optimization. It replaces the enumeration operation in subproblems with coarse or fine chaotic search. This can expand the search range in the early stage of iteration to avoid local convergence, and narrow the search range in the later stage of iteration to ensure optimization accuracy. This greatly improves computational efficiency, significantly reduces computation time, and improves the global solution speed and local optimization accuracy of POA. It is suitable for the efficient solution of large-scale cascade reservoir group joint scheduling optimization problems.

[0043] 2. Considering that the regularity, randomness and ergodicity of chaotic search are very suitable for improving the optimization process, chaotic search is applied to the two-stage optimization problem of POA. By using variable-scale chaotic search to traverse all states without repetition within the computational range, it can effectively deal with the large-scale reservoir group drawdown optimization scheduling problem and provide scientific support for the operation and control of cascade reservoir groups during the drawdown period. Attached Figure Description

[0044] Figure 1 This is a flowchart of the variable-scale chaotic search optimization method for drawdown scheduling of cascade reservoir groups according to the present invention;

[0045] Figure 2 This is a graph showing the calculated results of the Wudongde Reservoir under normal water conditions during the drawdown period;

[0046] Figure 3 This is a graph showing the calculated results of the Baihetan Reservoir under normal water conditions during the drawdown period;

[0047] Figure 4 This is a graph showing the calculated results for the Xiluodu Reservoir under normal water conditions during the drawdown period;

[0048] Figure 5 This is a diagram showing the calculated results of the Xiangjiaba Reservoir under normal water year conditions during the drawdown period. Detailed Implementation

[0049] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0050] like Figure 1 As shown, a variable-scale chaotic search optimization method for drawdown scheduling of a cascade reservoir group includes the following steps:

[0051] 1) Determine the initial calculation conditions, select the reservoir group participating in the joint drawdown scheduling, and clarify the objective function, constraints, and decision variables of the reservoir group drawdown scheduling. In this embodiment, the objective function of the reservoir group drawdown optimization scheduling can be described as follows: Given the initial water level, final water level, inflow process, and interval runoff process of the reservoir group participating in the joint scheduling during the drawdown period, and under the condition of meeting the various complex constraints such as water level and flow involved in multiple scheduling needs such as water supply, ecology, and navigation, determine the optimal stage water level operation process of the reservoir group to maximize the total power generation of the cascade reservoir group during the drawdown period, so as to give full play to the comprehensive utilization benefits of water resources. The objective function of the mathematical model for the optimization of cascade reservoir skirt drawdown scheduling is as follows:

[0052]

[0053] In the formula: F is the total power generation during the drawdown period, N is the number of reservoirs, i is the reservoir number, and i = 1, 2, ..., N, T is the total number of drawdown periods, j is the period number, and j = 1, 2, ..., T, A i Let Q be the output coefficient of reservoir i. i,j The power generation flow (m³) of reservoir i during time period j 3 / s), H i,jLet Δj be the average generating head (m) of reservoir i during time period j after deducting head loss, and let Δj be the time period length (h). To ensure the feasibility and availability of the optimization results, the reservoir group drawdown optimization scheduling problem needs to consider a large number of complex constraints, mainly including:

[0054] ① Hydraulic constraints:

[0055] ② Water balance constraints:

[0056] ③ Time-limited water level constraints:

[0057] ④ Discharge flow constraint:

[0058] ⑤ Power plant output constraints:

[0059] ⑥ Initial and final water level constraints: Z i,start =Z i,end

[0060] ⑦ System output constraints:

[0061] ⑧ Non-negativity constraint: All variables must be non-negative.

[0062] In the formula: I i,j and R i,j Reservoir i is respectively j Inbound flow rate during the period (m 3 / s) and interval inflow (m 3 / s); S i-1,j Let m be the discharge flow of the (i-1)th reservoir during time period j. 3 / s); V i,j For the reservoir in j Storage capacity at the end of the period (m) 3 );Z min i,j and Z max i,j Q represents the lowest and highest upstream water levels (m) of reservoir i during time period j; min i,j and Q max i,j The minimum power generation flow (m³) of reservoir i during time period j are respectively. 3 / s) and maximum power generation flow (m 3 / s); P min i,j and P max i,j Z represents the minimum and maximum power output (kW) of reservoir i during time period j, respectively; i,start and Zi,end NP represents the initial reservoir water level (m) during the drawdown period of reservoir i and the control water level (m) at the end of the drawdown period, respectively. j Let be the minimum power output (kW) of the system during time period j;

[0063] 2) Set the calculation parameters, including the maximum number of iterations M, the convergence accuracy ε, the total number of coarse-scale chaotic searches CH1 and the total number of fine-scale chaotic searches CH2 in the variable-scale chaotic search;

[0064] 3) Generate the initial water level process for the drawdown scheduling of the reservoir group based on human experience or conventional methods. And calculate the initial distance step length Δ = (Δ) for each reservoir. i,j ) N×T ,in Z i,j Let Δ be the water level of reservoir i during time period j. i,j For the distance of reservoir i from the time period j, and Let N be the upper and lower limits of the water level of reservoir i in time period j, respectively; N be the number of reservoirs; T be the number of time periods for scheduling; and K be the initial number of discrete nodes.

[0065] 4) Initialize the number of iterations m = 1;

[0066] 5) Decompose the reservoir group drawdown scheduling optimization problem of the current iteration number into T-1 two-stage stepwise optimization subproblems;

[0067] 6) For T-1 two-stage stepwise optimization subproblems, a coarse-scale chaotic search or a fine-scale chaotic search is randomly selected according to probability for optimization: when p > m / M, a coarse-scale chaotic search is executed; otherwise, a fine-scale chaotic search is executed, where p is a random number between 0 and 1.

[0068] 7) Initialize the time period j = 1;

[0069] 8) The water levels of each reservoir during time period j-1 are: Water levels of each reservoir during time period j+1 If p > m / M, for Perform coarse-scale chaotic search optimization; if p ≤ m / M, for Perform fine-scale chaotic search to find the optimal solution.

[0070] 9) Let j = j + 1. If j ≤ T - 1, go to step 8); otherwise, go to step 10.

[0071] 10) Calculate the scheduling process between two consecutive iterations. and like Then shrink all reservoirs away from the walking distance, making Proceed to step 11), otherwise proceed to step 5.

[0072] 11) Let m = m + 1. If m > M, then go to step 12); otherwise, go to step 5.

[0073] 12) Terminate the calculation and output the optimal water level process for each reservoir.

[0074] In step 8), the coarse-scale chaotic search optimization is performed, including the following steps:

[0075] 8.1) Normalization generates chaotic variables: the water levels of each reservoir in time period j are normalized. Normalization is performed to generate initial chaotic variables. The mapping is to the interval [0,1], where λ is the chaos control parameter. When λ = 4, it is in a completely chaotic state.

[0076] 8.2) Iterative generation of coarse-scale chaotic search variables: Based on the chaotic mapping generation method, the chaotic variables of the first coarse-scale chaotic search are generated. Chaotic variables in the first coarse-scale chaos search (ch1)

[0077] 8.3) Coarse-scale chaotic search optimization calculation: including the following steps:

[0078] 8.3.1) Initialize the number of coarse-scale chaotic searches ch1 = 1;

[0079] 8.3.2) Calculation The calculation method is as follows: Compare and like Superior make

[0080] 8.3.3) Let ch1 = ch1 + 1; if ch1 ≤ CH1, go to step 8.3.2); otherwise, go to step 9.

[0081] In step 8), a fine-scale chaotic search optimization is performed, including the following steps:

[0082] 8.4) Normalization generates chaotic variables: the water levels of each reservoir in time period j are normalized. Normalization is performed to generate initial chaotic variables. To map to the interval [0,1];

[0083] 8.5) Iterative Generation of Fine-Scale Chaotic Search Variables: Based on the chaotic mapping generation method, the chaotic variables of the first fine-scale chaotic search are... The chaotic variables in the second fine-scale chaotic search are:

[0084] 8.6) Fine-scale chaotic search optimization calculation: including the following steps:

[0085] 8.6.1) Initialize the number of coarse-scale chaotic searches ch2 = 1;

[0086] 8.6.2) will Inverse normalization to Interval, calculation The calculation method is as follows: In the formula, η is an adaptive parameter, which is a random number between 0 and 1. and like Superior make

[0087] 8.6.3) Let ch2 = ch2 + 1; if ch2 ≤ CH2, go to step 8.6.2); otherwise, go to step 9.

[0088] Based on the above method, a variable-scale chaotic search optimization system for drawdown scheduling of cascade reservoir groups is provided, which is used for the variable-scale chaotic search optimization method for drawdown scheduling of cascade reservoir groups, including:

[0089] 1) Initialization module: Determine the objective function, constraints and decision variables of the drawdown optimization scheduling of the cascade reservoir group, and generate the initial drawdown scheduling process of the reservoir group that meets the constraints based on human experience or conventional methods under various constraints;

[0090] 2) Stepwise optimization decomposition module: The reservoir group drawdown optimization scheduling problem of the current iteration number is decomposed into multiple two-stage stepwise optimization sub-problems;

[0091] 3) Variable-scale chaotic search selection module: It judges each stepwise optimization subproblem and randomly selects coarse-scale chaotic search or fine-scale chaotic search for optimization solution according to probability;

[0092] 4) Coarse-scale chaotic search module: Performs coarse-scale chaotic search, including normalization to generate chaotic variables, iterative generation of coarse-scale chaotic search variables, coarse-scale chaotic search optimization calculation, and stopping condition judgment functions, in order to obtain better optimization results;

[0093] 5) Fine-scale chaotic search module: Performs fine-scale chaotic search, including normalization to generate chaotic variables, iterative generation of fine-scale chaotic search variables, fine-scale chaotic search optimization calculation and stopping condition judgment functions, so as to obtain better optimization results;

[0094] 6) Output module: Repeatedly execute the stepwise optimization decomposition module, the variable-scale chaotic search selection module, the coarse-scale chaotic search module, and then the fine-scale chaotic search module until the preset iteration stopping condition is met, obtain the optimal scheduling result, and output it as the final cascade reservoir group drawdown optimization scheduling result.

[0095] This study takes the drawdown optimization scheduling problem of the cascade reservoir group in the lower reaches of the Jinsha River as an example. This basin is rich in hydropower resources and is a strategic water source and major hydropower development base for my country's water resource allocation, significantly influencing flood control and water resource utilization patterns in the upper reaches of the Yangtze River (Sichuan-Chongqing section) and the middle and lower reaches of the Yangtze River. The lower reaches of the Jinsha River have four large reservoirs—Wudongde, Baihetan, Xiluodu, and Xiangjiaba—with a total regulating capacity of 20.821 billion m³. 3 The total installed capacity is 46,400 MW. The drawdown period is a crucial time for the cascade reservoirs in the lower reaches of the Jinsha River to redistribute water volume through hydraulic compensation to regulate water levels during rainy and dry seasons. Drawdown period scheduling has two significant characteristics: first, it has a long duration, accounting for more than half of the total annual duration; second, it involves multiple scheduling demands, requiring comprehensive consideration of power generation, water supply, ecological, and navigation needs, making it a key link in maximizing comprehensive benefits.

[0096] The selected drawdown scheduling period is from January 1st to June 30th. The water levels of each reservoir at the beginning of the year are the normal storage levels, and the final drawdown levels are the flood control limits. Scheduling periods are divided into ten-day periods. Five typical drawdown periods were selected, representing five different inflow rates: abundant, slightly abundant, normal, slightly dry, and dry years. The method of this invention and the Power of Atmosphere (POA) were used to optimize the drawdown scheduling of the cascade reservoir group.

[0097] Table 1 lists the comparison of the method of the present invention and the POA calculation results when considering different numbers of power plants and different level years.

[0098] Table 1 Comparison of the method of the present invention and the POA calculation results.

[0099]

[0100]

[0101] As shown in Table 1, the power generation calculated by the method of this invention is similar to that of POA, but the computational advantages are as follows: ① From the comparison of power generation, both the method of this invention and POA continuously approach the global optimal solution, and the power generation of the method of this invention is greater than that of POA, indicating improved computational efficiency. Furthermore, the advantage becomes more significant as the scale of the reservoir group calculation increases. ② The computation time of the method of this invention is shorter than that of POA, approximately 33% of the computation time of POA. Moreover, the computational performance advantage becomes more prominent as the scale of the hydropower station calculation increases. Therefore, compared with POA, this invention can increase power generation, improve computational accuracy, reduce computation time, and increase solution speed, making it applicable to the optimal scheduling of giant cascade reservoir groups.

[0102] like Figures 2-5 The diagrams shown illustrate the changes in water level and power output of each reservoir. As can be seen from the diagrams, the optimized drawdown scheduling results for each reservoir satisfy all constraints. Wudongde and Baihetan reservoirs, being upstream in the cascade reservoir system and possessing greater regulating capacity, experience earlier drawdowns than downstream Xiluodu and Xiangjiaba reservoirs, resulting in less pre-flood storage. This allows downstream reservoirs to fully utilize the cascade head effect, increasing their power generation and thus the total power generation of the cascade reservoir group. Similar conclusions are drawn from the optimization results of the reservoir group under different drawdown periods and annual water flow conditions, which aligns with actual conditions. This demonstrates that the calculation results of the method presented in this invention are reasonable and reliable, and can be directly applied to the scheduling, operation, and production practices of the cascade reservoir group in the basin.

[0103] This invention presents a variable-scale chaotic search optimization method and system for drawdown scheduling of cascade reservoir groups. Based on a stepwise optimization method, it combines variable-scale chaotic search with a two-stage stepwise optimization approach. It replaces enumeration operations in subproblems with either coarse or fine chaotic search. This approach expands the search range in the early iterations to avoid local convergence and narrows the search range in the later iterations to ensure optimization accuracy. This significantly improves computational efficiency, reduces computation time, and enhances the global solution speed and local optimization accuracy of Power of Attention (POA). It is suitable for efficiently solving large-scale cascade reservoir group joint scheduling optimization problems. Considering the regularity, randomness, and ergodicity of chaotic search, which are highly suitable for improving the optimization process, chaotic search is applied to the two-stage optimization problem of POA. Variable-scale chaotic search traverses all states without repetition within the computational range, effectively addressing large-scale reservoir group drawdown optimization scheduling problems and providing scientific support for the operation and control of cascade reservoir groups during the drawdown period.

[0104] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A variable-scale chaotic search optimization method for drawdown scheduling of a cascade reservoir group, characterized in that: Includes the following steps: 1) Determine the initial calculation conditions, select the reservoir group to participate in the joint drawdown scheduling, and clarify the objective function, constraints, and decision variables of the reservoir group drawdown optimization scheduling; 2) Set the calculation parameters, including the maximum number of iterations. Convergence accuracy Total number of coarse-scale chaotic searches in variable-scale chaotic search Total number of fine-scale chaotic searches ; 3) Generate the initial water level process for the drawdown scheduling of the reservoir group based on human experience or conventional methods. And calculate the initial distance step length of each reservoir. ,in , For reservoir During the period water level, For reservoir During the period The distance from the long walk, and Reservoirs During the period Upper and lower limits of water level, For the number of reservoirs, The number of time periods to be scheduled. The initial number of discrete elements; 4) Initialize the number of iterations ; 5) Decompose the reservoir group drawdown scheduling optimization problem of the current iteration number into... The sub-problem is gradually optimized in two stages; 6) To The subproblem is solved through a two-stage stepwise optimization process, randomly selecting either a coarse-scale chaotic search or a fine-scale chaotic search based on probability: when Perform a coarse-scale chaotic search if the condition is met, otherwise perform a fine-scale chaotic search, where... A random number between 0 and 1; 7) Initialization period ; 8) Time period The water levels of each reservoir are Time period Water levels in various reservoirs ,like ,against Perform coarse-scale chaotic search for optimization, if ,against Perform fine-scale chaotic search to find the optimal solution. ; 9) Order ,like Proceed to step 8); otherwise, proceed to step 10. 10) Calculate the scheduling process between two consecutive iterations. and ,like This will shrink the distance between all reservoirs and the walking distance, causing (If you are already in the same position, proceed to step 11; otherwise, proceed to step 5). 11) Order ,like If the result is positive, proceed to step 12; otherwise, proceed to step 5. 12) Terminate the calculation and output the optimal water level process for each reservoir.

2. The variable-scale chaotic search optimization method for drawdown scheduling of cascade reservoir groups as described in claim 1, characterized in that: In step 8), the coarse-scale chaotic search optimization is performed, including the following steps: 8.1) Normalization generates chaotic variables: The time period... Water levels in various reservoirs Normalization is performed to generate initial chaotic variables. To map to the interval [0,1] For chaotic control parameters, when It was in a state of complete chaos at that time; 8.2) Iterative generation of coarse-scale chaotic search variables: Based on the chaotic mapping generation method, the chaotic variables of the first coarse-scale chaotic search are generated. , No. Chaotic variables in coarse-scale chaotic search ; 8.3) Coarse-scale chaotic search optimization calculation: including the following steps: 8.3.1) Initialize the number of coarse-scale chaotic search iterations ; 8.3.2) Calculation The calculation method is as follows: ,Compare and ,like Superior ,make ; 8.3.3) Order ;like (If not, proceed to step 8.3.2), otherwise proceed to step 9).

3. The variable-scale chaotic search optimization method for drawdown scheduling of cascade reservoir groups as described in claim 1, characterized in that: In step 8), the fine-scale chaotic search optimization is performed, including the following steps: 8.4) Normalization generates chaotic variables: The time period... Water levels in various reservoirs Normalization is performed to generate initial chaotic variables. , to map to the interval [0,1]; 8.5) Iterative Generation of Fine-Scale Chaotic Search Variables: Based on the chaotic mapping generation method, the chaotic variables of the first fine-scale chaotic search are... , No. The chaotic variable in the sub-fine-scale chaotic search is ; 8.6) Fine-scale chaotic search optimization calculation: including the following steps: 8.6.1) Initialize the number of fine-scale chaotic searches ; 8.6.2) Calculation The calculation method is as follows: In the formula For adaptive parameters, a random number between 0 and 1 is used for comparison. and ,like Superior ,make ; 8.6.3) Order ;like (If not, proceed to step 8.6.2); otherwise, proceed to step 9.

4. A variable-scale chaotic search optimization system for drawdown scheduling of a cascade reservoir group, characterized in that: The variable-scale chaotic search optimization method for drawdown scheduling of a cascade reservoir group as described in any one of claims 1 to 3 includes: 1) Initialization module: Determine the objective function, constraints and decision variables of the drawdown optimization scheduling of the cascade reservoir group, and generate the initial drawdown scheduling process of the reservoir group that meets the constraints based on human experience or conventional methods under various constraints; 2) Stepwise optimization decomposition module: The reservoir group drawdown optimization scheduling problem of the current iteration number is decomposed into multiple two-stage stepwise optimization sub-problems; 3) Variable-scale chaotic search selection module: It judges each stepwise optimization subproblem and randomly selects coarse-scale chaotic search or fine-scale chaotic search for optimization solution according to probability; 4) Coarse-scale chaotic search module: Performs coarse-scale chaotic search, including normalization to generate chaotic variables, iterative generation of coarse-scale chaotic search variables, coarse-scale chaotic search optimization calculation, and stopping condition judgment functions, in order to obtain better optimization results; 5) Fine-scale chaotic search module: Performs fine-scale chaotic search, including normalization to generate chaotic variables, iterative generation of fine-scale chaotic search variables, fine-scale chaotic search optimization calculation, and stopping condition judgment functions, in order to obtain better optimization results; 6) Output module: Repeatedly execute the stepwise optimization decomposition module, the variable-scale chaotic search selection module, the coarse-scale chaotic search module, and then the fine-scale chaotic search module until the preset iteration stopping condition is met, obtain the optimal scheduling result, and output it as the final cascade reservoir group drawdown optimization scheduling result.

Images