论文部分内容阅读
针对遗传算法中初始解分布不均以及易早熟等问题,采用均匀设计方法来生成均匀分布的初始解以及自组织映射算法通过高低维空间映射来改变个体基因从而增强局部搜索能力,提出了均匀自组织映射遗传算法,弥补了传统遗传算法中初始解的生成过于随机以及进化过程中易陷入局部解的不足,并将此改进算法在梯级水库的长期优化调度中进行了应用.通过实例计算表明,与遗传算法以及标准粒子群算法相比,此方法拥有更好的全局寻优能力,与动态规划算法结果相近,并且有着较快的计算速度,从而验证了此方法用于处理梯级水库的长期优化调度问题的可行性与合理性.
Aiming at the problems of uneven initial distribution and precocious maturation in genetic algorithm, uniform design method is used to generate uniform initial solution and self-organizing mapping algorithm is used to change individual genes through high and low dimensional space mapping to enhance local search ability. The algorithm of tissue-mapping genetic algorithm can make up for the shortcomings of the traditional genetic algorithm, such as the generation of initial solution is too random and the solution is easy to fall into local solution in evolutionary process, and the improved algorithm is applied in the long-term optimal scheduling of cascade reservoirs. Compared with genetic algorithm and standard particle swarm optimization algorithm, this method has better global optimization ability, which is similar to the result of dynamic programming algorithm and has a faster calculation speed, which verifies this method is suitable for long-term optimization of cascade reservoirs Feasibility and rationality of scheduling problem.