Cascade power station scheduling method and system based on improved random fractal algorithm

Abstract

The invention discloses a cascade hydropower station scheduling method and system based on an improved random fractal algorithm, and belongs to the technical field of cascade hydropower station power generation scheduling. Aiming at the defects that a traditional random fractal algorithm is insufficient in local optimization capacity when solving a cascade hydropower station power generation dispatching model and is prone to falling into a local optimal solution in the later stage, the following improvements are provided: (1) in the earlier stage of iteration, mutation operation of a differential evolution algorithm is introduced into a second updating stage of the random fractal algorithm, and the random fractal algorithm is used for solving the cascade hydropower station power generation dispatching model; the global optimization ability of the algorithm is improved, and the diversity of generated groups is ensured; (2) in the later stage of iteration, a first random individual of the mutation operation of the differential evolution algorithm in the step (1) is changed into a global optimal individual, so that a population is subjected to local fine search around the optimal individual, and the search precision of the algorithm is improved; and (3) adopting a self-adaptive mutation rate for mutation operation in the step (1) and the step (2), and increasing the possibility that the algorithm jumps out of a local optimal solution while adjusting the balance of global search and local search.

Description

technical field [0001] The invention belongs to the technical field of power generation scheduling of cascade hydropower stations, and more particularly relates to a cascade power station scheduling method and system based on an improved random fractal algorithm. Background technique [0002] The optimal scheduling of cascade reservoirs is a kind of high-dimensional, non-convex and nonlinear combinatorial optimization problem. Selecting an appropriate optimization algorithm to efficiently solve the scheduling model is the key to realizing optimal scheduling of cascade reservoirs. Traditional reservoir optimization and scheduling methods mainly include dynamic programming, successive approximation dynamic programming, and step-by-step optimization algorithms. However, these methods have problems such as "dimension disaster", slow convergence speed, and initial solution dependence. When the hydraulic constraints of the (PSO) and other methods use group-scale optimization and ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the defects and improvement needs of the prior art, the present invention provides a cascade power plant scheduling method and system based on the improved stochastic fractal algorithm, aiming to apply the new meta-heuristic algorithm - stochastic fractal search algorithm (SFSS) to reservoir power generation scheduling field, and overcome the shortcomings of the random fractal search algorithm in solving the cascade hydropower generation dispatching model, which is insufficient in local optimization ability and easy to fall into the local optimal solution in the later stage.

Images