作者: Li-Chiu Chang , Fi-John Chang
DOI: 10.1002/HYP.226
关键词:
摘要: This paper presents a new approach to improving real-time reservoir operation. The combines two major procedures: the genetic algorithm (GA) and adaptive network-based fuzzy inference system (ANFIS). GA is used search optimal operating histogram based on given inflow series, which can be recognized as base of input–output training patterns in next step. ANFIS then built create system, construct suitable structure parameters, estimate water release according depth situation. practicability effectiveness proposed tested operation Shihmen Taiwan. current M-5 rule curves are also evaluated. simulation results demonstrate that this approach, comparison with curves, has superior performance regard prediction total deficit generalized shortage index (GSI). Copyright © 2001 John Wiley & Sons, Ltd.