基于改进遗传算法的梯级电站日经济优化调度研究
Research on Daily Optimal Scheduling of Cascade Power Plant Based on Improved Genetic Algorithm
蔡望 1陈仁义 1彭辉1
作者信息
- 1. 三峡大学水利与环境学院,湖北 宜昌 443000
- 折叠
摘要
本文指出梯级水电站在优化运行时往往伴随着多约束、非线性、多阶段的组合优化问题,在进行求解时由于传统遗传算法等其他相似智能算法对水库优化调度模型求解时局部搜索能力较差、在进化后期搜索效率较低,往往容易产生早熟收敛的问题.针对此问题,本文对遗传算法进行适当改进,在分时电价背景下,将该算法应用于溪洛渡-向家坝梯级电站日经济效益最大为目标的日优化调度模型中.通过将改进遗传算法与传统遗传算法以及其他优化算法所得结果进行比较,发现改进遗传算法收敛较快、稳定性较好,克服了其他算法早收敛、陷入局部最优的缺点,可以为水库优化调度提供了一定参考价值.
Abstract
This paper points out that the optimization operation of cascade hydropower stations is often accompanied by multi constraint,nonlinear,and multi-stage combination optimization problems.When solving,traditional genetic algorithms and other similar intelligent algorithms have poor local search ability and low search efficiency in the later stage of evolution,which often leads to premature convergence problems.In response to this issue,this paper makes appropriate improvements to the genetic algorithm and applies it to the daily optimization scheduling model of the Xiluodu-Xiangjiaba cascade power station with the goal of maximizing daily economic benefits in the context of time of use electricity prices.By comparing the results obtained by the improved genetic algorithm with traditional genetic algorithm and other optimization algorithms,it is found that the improved genetic algorithm converges quickly and has good stability,overcoming the shortcomings of other algorithms such as early convergence and falling into local optima.This can provide certain reference value for reservoir optimization scheduling.
关键词
梯级水电站/改进遗传算法/日经济效益/短期日优化调度Key words
cascade hydropower station/improved genetic algorithm/daily economic benefits/short-term daily optimized scheduling引用本文复制引用
基金项目
国家自然科学基金(51379108)
重点研发计划专项(2022YFC3005603-01)
出版年
2024
NETL
NSTL
万方数据