Cascade reservoir group optimization scheduling parallel variable parameter simulated annealing optimization method

Through the parallel variable parameter simulation annealing optimization method in cascade reservoir group optimization scheduling, the simulation annealing algorithm and multi-core parallel computing are used to solve the problems of insufficient search capabilities and low computing efficiency in large-scale reservoir group optimization scheduling, and the efficient global maximum is achieved Improved optimization solution and calculation accuracy.

CN119005581AActive Publication Date: 2024-11-22CHANGJIANG SURVEY PLANNING DESIGN & RES CO LTD

Patent Information

Application Number
CN202411019942.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-22
Estimated Expiration
2044-07-29

AI Technical Summary

Technical Problem

The optimization and scheduling problem of cascade reservoir clusters is a large-scale, high-dimensional, multi-stage, strong constraints, and nonlinear optimization problem. Traditional methods such as dynamic programming, genetic algorithms and particle cluster algorithms have insufficient computational efficiency and accuracy, especially in large When optimizing the scheduling of large-scale reservoir groups, it is easy to fall into dimensional disasters and premature convergence.

Method used

The parallel variable parameter simulation annealing optimization method is adopted, and searches in multi-core parallel computing through simulated annealing algorithm, and iterative optimization is used to use different annealing control parameters to gradually approximate the global optimal solution. The initial trajectory is generated in combination with conventional dynamic programming methods and in multi-threading Perform annealing search.

Benefits of technology

It effectively overcomes the problems of insufficient search capabilities and low computing speed of traditional methods, significantly improves the calculation efficiency and accuracy of optimization scheduling of cascade reservoir groups, reduces calculation time and memory usage, avoids dimensional disasters, and improves global search capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119005581A_ABST
    Figure CN119005581A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of cascade reservoir group optimization scheduling, and discloses a cascade reservoir group optimization scheduling parallel variable parameter simulated annealing optimization method comprising the following steps: determining initial calculation conditions including a target function and a constraint condition of cascade reservoir group optimization scheduling; setting calculation parameters; adopting a conventional dynamic planning method or an artificial experience decision to generate an initial trajectory of each reservoir meeting each constraint condition; and starting iterative computation, performing multi-core parallel computing search by adopting a simulated annealing method, and selecting different annealing control parameters in each thread to perform annealing search in different forms and repeated iterative optimization so as to gradually approach a global optimal solution and output a final optimal trajectory. According to the cascade reservoir group optimization scheduling parallel variable parameter simulated annealing optimization method, the problems of insufficient search capability and low calculation rate existing in large-scale hydropower system optimization scheduling solution of DDDP are effectively solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of optimal dispatching of a cascade reservoir group, and in particular to a parallel variable parameter simulated annealing optimization method for optimal dispatching of a cascade reservoir group. Background Art

[0002] As a renewable and clean energy, hydropower occupies an extremely important position in the history of my country's energy development and strongly supports the sustainable development of the economy and society. In recent years, my country's hydropower industry has developed rapidly, and the hydropower resources in major river basins have been further developed and utilized on a large scale, gradually realizing the cascade rolling development of hydropower basins and the optimal allocation of hydropower resources. The optimal dispatching of cascade reservoirs has significant comprehensive benefits in social, economic and ecological aspects, and can effectively promote the efficient utilization of water resources in the river basin and ensure the safe, stable and reliable operation of the power grid system. It has always been an important engineering practice and theoretical research topic in the field of hydropower energy systems. However, with the increasing scale of cascade reservoir systems, the dispatching problem has become more and more complex, resulting in an exponential increase in the calculation time and memory usage during the optimal dispatching process. The dimensionality disaster problem has become a scientific problem that must be faced by the optimal dispatching of reservoir groups. Therefore, actively developing efficient and practical methods that can adapt to the current situation to solve the joint optimal dispatching problem of cascade reservoir groups and shorten the gap with the actual dispatching situation has important engineering practical value for the dispatching and operation management of cascade reservoir groups.

[0003] From a mathematical point of view, the optimal scheduling problem of cascade reservoirs is a large-scale, high-dimensional, multi-stage, strongly constrained, and nonlinear optimization problem. The traditional dynamic programming method (DP) can solve the optimal scheduling problem of a single reservoir well, but it is difficult to avoid the curse of dimensionality when solving the optimal scheduling problem of cascade reservoirs. Although the operation methods of intelligent optimization algorithms such as genetic algorithms, differential evolution algorithms, and particle swarm algorithms are different when solving the optimal scheduling problem of reservoir groups, they all have defects such as low computational efficiency, high computational overhead, and insufficient convergence ability. Therefore, it is urgent to combine the latest achievements of computational science to develop some new and effective optimization methods to achieve the simultaneous reduction of memory usage and computational time, effectively alleviate the curse of dimensionality, and ensure the computational efficiency and solution accuracy of the optimal scheduling of large-scale reservoir groups.

[0004] Discrete differential dynamic programming (DDDP) is an improved dynamic programming method based on the theory of successive asymptotic approximation. The basic idea is: first, obtain the initial trajectory that meets various complex constraints based on experience or other methods; then, discretize the state variables of each reservoir at different time periods in the neighborhood of the trajectory to form a corridor; secondly, optimize between the discrete state combinations of each time period based on the conventional dynamic programming method, find a new optimal trajectory as the trial trajectory for the next iteration, and iterate repeatedly until the convergence condition is met. However, DDDP still has defects such as dimensionality curse, premature convergence, and low search ability when dealing with the problem of optimal scheduling of large-scale reservoir groups. Therefore, it is necessary to combine new technical means to improve the calculation mechanism of DDDP to improve the calculation efficiency and solution accuracy of solving the problem of optimal scheduling of reservoir groups.

[0005] Simulated annealing is a random optimization algorithm based on the Monte Carlo random simulation iterative solution strategy. Its basic idea is to start from a relatively high initial temperature, and with the continuous decrease of the temperature parameter, it combines the probability jump characteristics to randomly search for a global optimal solution in the feasible space. It has the advantages of jumping out of the local optimum and strong search ability. It has been successfully applied to control engineering, neural networks, signal processing and other fields. Parallel computing is a novel computer technology in recent years. Its essence is to make full use of the computing resources of multi-core processors and execute given computing operations in parallel on different threads through reasonable and feasible parallel computing modes. Summary of the invention

[0006] The purpose of the present invention is to address the deficiencies of the above-mentioned technologies and provide a parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoir groups, which effectively solves the problems of insufficient search capability and low calculation rate of DDDP in solving the optimal scheduling of large-scale hydropower systems.

[0007] To achieve the above object, the present invention provides a parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoirs, comprising the following steps:

[0008] 1) Determine the initial calculation conditions, including the objective function and constraints for the optimal operation of cascade reservoirs;

[0009] 2) Set calculation parameters;

[0010] 3) Generate the initial trajectory of each reservoir that meets all constraints by using conventional dynamic programming methods or artificial experience decision-making;

[0011] 4) Start iterative calculation, use simulated annealing method to perform multi-core parallel computing search, and select different annealing control parameters in each thread to perform different forms of annealing search, repeatedly iterate to find the best solution, gradually approach the global optimal solution, and output the final optimal trajectory.

[0012] Preferably, in step 2), the calculation parameters include the number of reservoirs N, the number of scheduling stages T, the maximum number of iterations M, the maximum number of annealing searches K, the number of parallel computing threads C, the convergence accuracy ε and the initial temperature parameter T of thread c 0,c and annealing control parameter y 0,c , where c(1≤c≤C).

[0013] Preferably, in step 3), conventional dynamic programming methods or manual experience decision-making are used to generate the initial trajectory of each reservoir that meets each constraint condition. And calculate the initial discrete step length of each reservoir Δ=(Δ i,j ) N×T , the calculation formula is as follows:

[0014]

[0015] in, represents the initial trajectory of reservoir i at stage j, Δ i,j represents the initial discrete step length of reservoir i in stage j, represents the upper limit of the water level of reservoir i at stage j, Z i,j represents the lower limit of the water level of reservoir i at stage j, and L is the initial discrete number.

[0016] Preferably, the step 4) comprises the following steps:

[0017] 4.1) Set the number of iterations m = 1,

[0018] 4.2) Set the number of stages j = 1;

[0019] 4.3) The current temperature parameter of thread c is T m,c =T 0,c ×(y 0,c ) m , perform parallel simulated annealing search to obtain the best trajectory;

[0020] 4.4) Calculate the water level difference between two consecutive optimal trajectories in each period. If Indicates that the current trajectory is optimal, ends this iteration, shrinks all discrete steps, and makes Go to step 45), otherwise go to step 42), and continue based on Perform parallel variable parameter simulated annealing searches;

[0021] 4.5) Let m=m+1. If m>M, go to step 46); otherwise, go to step 42).

[0022] 4.6) Stop calculation and output the final optimal trajectory.

[0023] Preferably, in step 4.3), performing parallel simulated annealing search comprises the following steps:

[0024] 4.3.1) The current trajectory Copy C times and distribute them to each thread for calculation. The current trajectory of thread c is

[0025] 4.3.2) Set the number of annealing operations k for thread c c =1;

[0026] 4.3.3) Combined with the current trajectory of thread c Randomly generate new annealing trajectories The calculation formula is as follows:

[0027]

[0028] Where η is the disturbance amplitude control parameter, and λ is a random number between -1 and 1;

[0029] 4.3.4) Calculate the current trajectory separately and new trajectories The scheduling result is and The following three situations are handled:

[0030] if Indicates new trajectory Better than current trajectory Then replace the current trajectory with the new trajectory, that is

[0031] if and In the formula, r is a random number between 0 and 1. In order to maintain the diversity of annealing search, the current trajectory is replaced by a new trajectory, that is,

[0032] If neither of the above two conditions is satisfied, No treatment;

[0033] 4.3.5) Let k c =k c +1, if k c ≤K, go to step 4.3.3); otherwise, go to step 4.3.6);

[0034] 4.3.6) After all threads have finished calculating, the trajectories of all C threads are The best trajectory is selected by comparison have Then the best trajectory obtained by implementing parallel variable parameter simulated annealing in stage j replaces the current trajectory, that is,

[0035] 4.3.7) Let j = j + 1. If j ≤ T, go to step 4.3), otherwise go to step 4.4).

[0036] Preferably, in step 4.3.3), the value of η is 0.5.

[0037] Compared with the prior art, the present invention has the following advantages:

[0038] 1. Simple and convenient operation, with natural parallel computing conditions, good optimization ability and strong robustness;

[0039] 2. Based on the simulated annealing strategy, the optimization trajectory is found through the cooling process of simulated annealing, which effectively avoids the comprehensive combination between the discrete states of each reservoir at each stage, overcomes the dimensionality curse problem of traditional methods, and improves the global search capability;

[0040] 3. Based on parallel computing means, the simulated annealing cooling process with different parameters is implemented in multiple threads to reduce the uncertainty of setting the parameters of a single simulated annealing operation. The parallel coupling calculation of different simulated annealing parameters is cleverly realized, further improving the calculation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of the parallel variable parameter simulated annealing optimization method for optimizing the scheduling of cascade reservoir groups of the present invention;

[0042] Figure 2 : is a calculation result diagram of the WDD reservoir under the flood condition in the wet year in the embodiment;

[0043] Figure 3 is a diagram showing the calculation results of the BHT reservoir under the flood conditions in the wet year in the embodiment;

[0044] Figure 4 : is a calculation result diagram of the XLD reservoir under the flood condition in the wet year in the embodiment;

[0045] Figure 5 Graph showing the calculation results of the XJB reservoir under flood conditions in a flood year in the embodiment. DETAILED DESCRIPTION

[0046] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] like Figure 1 As shown, a parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoir groups includes the following steps:

[0048] 1) Determine the initial calculation conditions, including the objective function and constraints of the optimal dispatching of cascade reservoirs. The optimal dispatching model of cascade reservoirs can be described as: given the initial water level, final water level, inflow runoff process and interval runoff process of each reservoir during the dispatching period, and under the condition of satisfying various complex constraints such as the corresponding water level, flow, and output of the reservoir group, by determining the optimal stage water level operation process of the reservoir group, the water energy resources are utilized to the maximum extent, and the total power generation of the reservoir group during the dispatching period is maximized. The objective function of the optimal dispatching of cascade reservoirs is as follows:

[0049]

[0050] Where: F is the total power generation during the dispatch period, N is the number of reservoirs, i is the reservoir number, and i=1 , 2,…,N, T is the total number of stages in the scheduling period, j is the stage number, and j = 1, 2,…,T, A i is the output coefficient of reservoir i, Q i,j is the power generation flow of reservoir i in stage j (m 3 / s), H i,j is the average power generation head of reservoir i in stage j after deducting the head loss (m), Δj is the stage length (h). To ensure the feasibility and availability of the optimization results, the hydropower scheduling problem needs to consider a large number of complex constraints, mainly including:

[0051] Hydraulic connection constraints:

[0052] Water balance constraints:

[0053] Time period water level constraints:

[0054] Outbound traffic constraints:

[0055] Power station output constraints:

[0056] Initial and final water level constraints: Z i,start =Z i,end

[0057] System output constraints:

[0058] Non-negative constraints: All variables are non-negative values;

[0059] Where: I i,j and R i,j are the inflow of reservoir i at stage j (m 3 / s) and interval inflow (m 3 / s), S i-1,jis the water discharge of the i-1th reservoir at stage j (m 3 / s), V i,j is the storage capacity of the reservoir at the end of stage j (m 3 ), Z min i,j and Z max i,j are the lowest and highest water levels before the dam of reservoir i at stage j (m), Q min i,j and Q max i,j are the minimum power generation flow of reservoir i in stage j (m 3 / s) and maximum power generation flow (m 3 / s), P min i,j and P max i,j are the minimum output (kW) and maximum output (kW) of reservoir i at stage j, Z i,start and Z i,end are the initial reservoir water level (m) and the controlled water level (m) at the end of the dispatch period of reservoir i, respectively. j is the minimum output of the system in stage j (kW);

[0060] By analyzing the above objective functions and constraints, it can be seen that the optimal dispatching of cascade reservoirs involves many complex constraints such as the water level, storage capacity, and flow of each reservoir. It is a large-scale, high-dimensional, multi-stage, strongly constrained, and nonlinear complex coupled optimal control problem. When DDDP optimization is used to solve this problem, as the number of reservoirs and the number of discretes continue to increase, the amount of calculation increases exponentially, there is a dimensionality curse, and it is easy to fall into local convergence;

[0061] 2) Set the calculation parameters, including the number of reservoirs N, the number of scheduling stages T, the maximum number of iterations M, the maximum number of annealing searches K, the number of parallel computing threads C, the convergence accuracy ε, and the initial temperature parameter T of thread c 0,c and annealing control parameter y 0,c , where c(1≤c≤C);

[0062] 3) Use conventional dynamic programming methods or manual experience decision-making to generate the initial trajectory of each reservoir that meets all constraints And calculate the initial discrete step length of each reservoir Δ=(Δ i,j ) N×T , the calculation formula is as follows:

[0063]

[0064] in, represents the initial trajectory of reservoir i at stage j, Δi,j represents the initial discrete step length of reservoir i in stage j, represents the upper limit of the water level of reservoir i at stage j, Z i,j represents the lower limit of the water level of reservoir i at stage j, and L is the initial discrete number;

[0065] 4) Start iterative calculation, use simulated annealing method to perform multi-core parallel computing search, and select different annealing control parameters in each thread to perform different forms of annealing search, repeatedly iterate to find the best solution, gradually approach the global optimal solution, and output the final optimal trajectory. The specific steps are as follows:

[0066] 4.1) Set the number of iterations m = 1,

[0067] 4.2) Set the number of stages j = 1;

[0068] 4.3) The current temperature parameter of thread c is T m,c =T 0,c ×(y 0,c ) m , perform parallel simulated annealing search to obtain the best trajectory. The parallel simulated annealing search includes the following steps:

[0069] 4.3.1) The current trajectory Copy C times and distribute them to each thread for calculation. The current trajectory of thread c is

[0070] 4.3.2) Set the number of annealing operations k for thread c c =1;

[0071] 4.3.3) Combined with the current trajectory of thread c Randomly generate new annealing trajectories The calculation formula is as follows:

[0072]

[0073] In the formula, η is the disturbance amplitude control parameter, η is generally taken as 0.5, and λ is a random number between -1 and 1;

[0074] 4.3.4) Calculate the current trajectory separately and new trajectories The scheduling result is and The following three situations are handled:

[0075] if Indicates new trajectory Better than current trajectory Then replace the current trajectory with the new trajectory, that is

[0076] if and In the formula, r is a random number between 0 and 1. In order to maintain the diversity of annealing search, the current trajectory is replaced by a new trajectory, that is,

[0077] If neither of the above two conditions is satisfied, No treatment;

[0078] 4.3.5) Let k c =k c +1, if k c ≤K, go to step 4.3.3); otherwise, go to step 4.3.6);

[0079] 4.3.6) After all threads have finished calculating, the trajectories of all C threads are The best trajectory is selected by comparison have Then the best trajectory obtained by implementing parallel variable parameter simulated annealing in stage j replaces the current trajectory, that is,

[0080] 4.3.7) Let j = j + 1, if j ≤ T, go to step 4.3), otherwise go to step 4.4);

[0081] 4.4) Calculate the water level difference between two consecutive optimal trajectories in each period. If Indicates that the current trajectory is optimal, ends this iteration, shrinks all discrete steps, and makes Go to step 45), otherwise go to step 42), and continue based on Perform parallel variable parameter simulated annealing searches;

[0082] 4.5) Let m=m+1. If m>M, go to step 46); otherwise, go to step 42).

[0083] 4.6) Stop calculation and output the final optimal trajectory.

[0084] Based on the recognition of the optimization scheduling problem of cascade reservoir groups and the DDDP solution process, the present invention combines simulated annealing method, parallel computing and DDDP, takes DDDP as the basic execution framework, and first obtains the initial scheduling state that meets the constraints; then, in the iterative process, for the optimization scheduling problem at each stage, the simulated annealing method is used to perform multi-core parallel computing search, and different annealing control parameters are selected in each thread to perform different forms of annealing search; finally, iterative optimization is repeatedly performed to gradually approach the global optimal solution. The dual advantages of simulated annealing strategy that can quickly jump out of local search and parallel computing technology that can implement different mode calculations are fully utilized, and parallel variable parameter simulated annealing search is performed at each stage, and the optimization trajectory is constantly sought for iterative update to effectively overcome the problem of insufficient search ability of DDDP.

[0085] This paper takes the optimization dispatching problem of a cascade reservoir group in the upper reaches of the Yangtze River as an example for research. The basin is rich in hydropower resources and is a strategic water source for my country's water resources allocation and a major base for hydropower development. It significantly affects the flood control and water resource utilization pattern in the Sichuan-Chongqing section of the upper reaches of the Yangtze River and the middle and lower reaches of the Yangtze River. Four large reservoirs, WDD, BHT, XLD and XJB, have been built in the basin, which undertake the comprehensive utilization of water resources such as flood control, power generation, shipping, water supply, and ecology. The total regulating storage capacity of the four reservoirs is 20.821 billion m 3 , with a total installed capacity of 46,400MW. In order to respond to the national call for energy conservation and emission reduction, give full play to the scale benefits and cascade compensation benefits of river basin reservoirs, and implement the optimal dispatching of cascade reservoirs, it is very important.

[0086] To this end, three different water inflows, namely, a dry year (frequency 75%), a normal year (frequency 50%) and a wet year (frequency 25%), were selected, and the method of the present invention was applied to the optimal scheduling of this cascade reservoir group to verify the rationality, feasibility, high efficiency and practicality of the method of the present invention in actual engineering applications.

[0087] Table 1 Comparison of the calculation results of the method of the present invention and DDDP

[0088]

[0089]

[0090] As shown in Table 1, compared with DDDP, the computational advantages of the method of the present invention are as follows:

[0091] 1. From the comparison of power generation, the method of the present invention and DDDP are constantly approaching the global optimal solution, and the power generation of the method of the present invention is greater than that of DDDP, indicating that the search ability is stronger, and as the calculation scale increases, its advantages in avoiding the curse of dimensionality and escaping the local optimal ability become more significant;

[0092] 2. The calculation time of the method of the present invention is significantly less than that of DDDP, which is only about 4% of the calculation time of DDDP. As the scale of the reservoir group and the calculation stage increase, the calculation performance advantage will become more prominent. It can be seen that compared with DDDP, the present invention can greatly reduce the calculation time and significantly improve the calculation efficiency, both in terms of calculation results and calculation time. It has excellent characteristics such as fast calculation efficiency and high calculation accuracy, and provides an effective method for the optimization and dispatching of giant cascade reservoir groups.

[0093] Figure 2 to Figure 5 The schematic diagrams of the water level and output change process of each reservoir in a normal water year are listed respectively. The optimized dispatching results of each reservoir meet all the constraints, effectively realizing the full-cycle dispatching process such as flood control and security in the flood season, rapid filling in the water storage period, and gradual decline in the dry season, giving full play to the comprehensive benefits of the head and water volume of the reservoir group, indicating that the calculation results of the method of the present invention are reasonable and effective, and can provide technical support for the actual dispatching and operation management of the cascade reservoir group.

[0094] The parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoir groups of the present invention is simple to operate, convenient and easy to implement, has natural parallel computing conditions, good optimization ability and strong robustness; based on the simulated annealing strategy, the optimization trajectory is found through the cooling process of simulated annealing, which effectively avoids the comprehensive combination between the discrete states of each reservoir at each stage, overcomes the dimensionality curse problem of traditional methods, and improves the global search capability; based on parallel computing means, the simulated annealing cooling process with different parameters is implemented in multiple threads to reduce the uncertainty of the setting of a single simulated annealing operation parameter, and cleverly realizes the parallel coupling calculation of different simulated annealing parameters, further improving the calculation accuracy.

[0095] It will be easily understood by those skilled in the art that the above description is only 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 in the protection scope of the present invention.

Claims

1. A parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoirs, characterized by: The steps include: 1) Determine the initial calculation conditions, including the objective function and constraints for the optimal operation of cascade reservoirs; 2) Set calculation parameters; 3) Generate the initial trajectory of each reservoir that meets all constraints by using conventional dynamic programming methods or artificial experience decision-making; 4) Start iterative calculation, use simulated annealing method to perform multi-core parallel computing search, and select different annealing control parameters in each thread to perform different forms of annealing search, repeatedly iterate to find the best solution, gradually approach the global optimal solution, and output the final optimal trajectory.

2. The parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoirs according to claim 1, characterized in that: In step 2), the calculation parameters include the number of reservoirs N, the number of scheduling stages T, the maximum number of iterations M, the maximum number of annealing searches K, the number of parallel computing threads C, the convergence accuracy ε and the initial temperature parameter T of thread c. 0,c and annealing control parameter y 0,c , where c(1≤c≤C).

3. The parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoirs as claimed in claim 2, characterized in that: In step 3), conventional dynamic programming methods or manual experience decision-making are used to generate initial trajectories of each reservoir that meet various constraints. And calculate the initial discrete step length of each reservoir Δ=(Δ i,j ) N×T , the calculation formula is as follows: in, represents the initial trajectory of reservoir i at stage j, Δ i,j represents the initial discrete step length of reservoir i in stage j, represents the upper limit of the water level of reservoir i at stage j, Z i,j represents the lower limit of the water level of reservoir i at stage j, and L is the initial discrete number.

4. The parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoirs as claimed in claim 3, characterized in that: The step 4) includes the following steps: 4.1) Set the number of iterations m = 1, 4.2) Set the number of stages j = 1; 4.3) The current temperature parameter of thread c is T m,c =T 0,c ×(y 0,c ) m , perform parallel simulated annealing search to obtain the best trajectory; 4.4) Calculate the water level difference between two consecutive optimal trajectories in each period. If Indicates that the current trajectory is optimal, ends this iteration, shrinks all discrete steps, and makes Go to step 45), otherwise go to step 42), and continue based on Perform parallel variable parameter simulated annealing searches; 4.5) Let m=m+1. If m>M, go to step 46); otherwise, go to step 42). 4.6) Stop calculation and output the final optimal trajectory.

5. The parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoirs as claimed in claim 4, characterized in that: In the step 4.3), performing parallel simulated annealing search comprises the following steps: 4.3.1) The current trajectory Copy C times and distribute them to each thread for calculation. The current trajectory of thread c is 4.3.2) Set the number of annealing operations k for thread c c =1; 4.3.3) Combined with the current trajectory of thread c Randomly generate new annealing trajectories The calculation formula is as follows: Where η is the disturbance amplitude control parameter, and λ is a random number between -1 and 1; 4.3.4) Calculate the current trajectory separately and new trajectories The scheduling result is and The following three situations are handled: if Indicates new trajectory Better than current trajectory Then replace the current trajectory with the new trajectory, that is if and In the formula, r is a random number between 0 and 1. In order to maintain the diversity of annealing search, the current trajectory is replaced by a new trajectory, that is, If neither of the above two conditions is satisfied, No treatment; 4.3.5) Let k c =k c +1, if k c ≤K, go to step 4.3.3); otherwise, go to step 4.3.6); 4.3.6) After all threads have finished calculating, the trajectories of all C threads are The best trajectory is selected by comparison have Then the best trajectory obtained by implementing parallel variable parameter simulated annealing in stage j replaces the current trajectory, that is, 4.3.7) Let j = j + 1. If j ≤ T, go to step 4.3), otherwise go to step 4.4).

6. The parallel variable parameter simulated annealing optimization method for optimal scheduling of cascade reservoirs as claimed in claim 5, characterized in that: In the step 4.3.3), the value of η is 0.5.

Citation Information

Patent Citations

  • Multi-core parallel simulated annealing method based on thread constructing module

    CN101593132A

  • Multi-core adaptive & parallel simulated annealing genetic algorithm based on cloud controller

    CN101826167A

  • Genetic simulated annealing method for solving new words in Chinese segmentation

    CN105740227A

  • Cascade reservoir group optimized dispatching multi-core parallel computing design method based on Fork / Join framework

    CN107015861A

  • Parallel simulated annealing quick optimization method for inversion measurement of multiple thermophysical parameters of material

    CN111898237A

Cited By

  • Cascade reservoir water storage period combined power generation optimization scheduling method and system and storage medium

    CN119692723A

  • Cascade reservoir water storage period joint power generation optimization scheduling method, system and storage medium

    CN119692723B

  • Water engineering dispatching intelligent parallel caching system for intelligent water conservancy and optimization method

    CN121070820A

  • Intelligent parallel cache system and optimization method for water conservancy project scheduling oriented to smart water conservancy

    CN121070820B

  • Multi-objective optimization reservoir gate strategy recommendation method and system

    CN121599196A