Multi-objective evolutionary algorithm for operating parallel reservoir system
Multi-objective evolutionary algorithm for operating parallel reservoir system
复制标题
DOI：
10.1016/j.jhydrol.2009.07.061
复制
发表时间：
2009-10
影响因子：
6.4
通讯作者：
Li-Chiu Chang;F. Chang
中科院分区：
文献类型：
--
作者：
Li-Chiu Chang;F. Chang
本文应用一种多目标演化演算法--非劣排序遗传演算法（NSGA-II），来研究台湾多水库系统的运作。翡翠、石门水库是台湾北方最重要的供水水库，供应七百多万居民的生活及工业用水。每日运行模拟模型的开发，以指导水库系统的释放，然后计算短缺指数（SI）的两个水库在一个长期的模拟期。NSGA-II用于通过识别最佳联合操作策略来最小化SI值。基于一个49 year的数据集，我们证明了更好的操作策略将减少短缺指数为两个水库。结果表明，NSGA-II提供了一个有前途的方法。帕累托前沿最优解确定了两个水库的运营折衷方案，预计将改善联合运营。
This paper applies a multi-objective evolutionary algorithm, the non-dominated sorting genetic algorithm (NSGA-II), to examine the operations of a multi-reservoir system in Taiwan. The Feitsui and Shihmen reservoirs are the most important water supply reservoirs in Northern Taiwan supplying the domestic and industrial water supply needs for over 7 million residents. A daily operational simulation model is developed to guide the releases of the reservoir system and then to calculate the shortage indices (SI) of both reservoirs over a long-term simulation period. The NSGA-II is used to minimize the SI values through identification of optimal joint operating strategies. Based on a 49year data set, we demonstrate that better operational strategies would reduce shortage indices for both reservoirs. The results indicate that the NSGA-II provides a promising approach. The pareto-front optimal solutions identified operational compromises for the two reservoirs that would be expected to improve joint operations.