论文部分内容阅读
在使用传统优化方法处理梯级电站数量庞大的长期优化调度时会出现“维数灾”及寻优效果差等问题。研究工作利用遗传算法进行计算,并对其进行了利于计算的改进,将自适应的控制理论加入交叉和变异算子,让其根据适应度的值自动改变,生成初始群体时使用混沌理论,并引入模拟退火方法,将两者的优点结合起来,生成了改进模拟退火遗传算法,提升了全局寻优能力和局部搜索能力,避免了算法陷入局部最优解。将改进后的算法程序应用于建立的模型中,通过与常规遗传算法的比较与分析,结果表明改进模拟退火遗传算法全局搜索能力强、求解效果好,为解决梯级水电站长期优化调度提供了新方法。
When using the traditional optimization method to deal with the long-term optimal scheduling of a large number of cascaded hydropower stations, there will be problems such as “dimensionality” and poor optimization. The research uses genetic algorithm to calculate and make some improvements to the calculation. The adaptive control theory is added to the crossover and mutation operator to make it change automatically according to the fitness value. The chaos theory is used to generate the initial population. The simulated annealing method is introduced to combine the advantages of the two. The improved simulated annealing genetic algorithm is generated, which improves the global search ability and local search ability, and avoids the algorithm falling into the local optimal solution. The improved algorithm is applied to the established model. Compared with the conventional genetic algorithm, the results show that the improved global search ability of simulated annealing genetic algorithm is good and the solution effect is good, which provides a new method to solve the long-term optimal scheduling of cascaded hydropower stations .