Premium
Linking genetic algorithms with stochastic dynamic programming to the long‐term operation of a multireservoir system
Author(s) -
Huang WenCheng,
Yuan LunChin,
Lee ChiMing
Publication year - 2002
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/2001wr001122
Subject(s) - curse of dimensionality , term (time) , genetic algorithm , dynamic programming , computer science , mathematical optimization , stochastic programming , genetic programming , algorithm , mathematics , artificial intelligence , physics , quantum mechanics
The objective of this paper is to present a genetic algorithm‐based stochastic dynamic programming (GA‐based SDP) to cope with the dimensionality problem of a multiple‐reservoir system. The joint long‐term operation of a parallel reservoir system in the Feitsui and Shihmen reservoirs in northern Taiwan demonstrates the successful application of the proposed GA‐based SDP model. Within the case study system it is believed that GA is a useful technique in supporting optimization. Though the employment of GA‐based SDP may be time consuming as it proceeds through generation by generation, the model can overcome the “dimensionality curse” in searching solutions. Simulation results show Feitsui's surplus water can be utilized efficiently to fill Shihmen's deficit water without affecting Feitsui's main purpose as Taipei city's water supply. The optimal joint operation suggests that Feitsui, on average, can provide 650,000 m 3 /day and 920,000 m 3 /day to Shihmen during the wet season and dry season, respectively.