Enhancing Cellular Automata via Tabu Search for Optimal Operation of Hydropower Systems

One of the environmentally-friendly solutions to meeting energy consumption is multi-reservoir hydropower systems. The operation of a multi-reservoir hydropower system is an entirely complex problem due to a wide range of decision variables. Classic algorithms often get stuck in the local optimum and cannot successfully address these problems. Modern algorithms are more effective than classic ones, although their computational time is very high. In this study, an innovative hybrid model is proposed, called cellular automata-tabu search (CA-TS) to optimally operate multi-reservoir systems. For simplification, CA divides the problem into several sub-problems, which the number of them is the same as the length of operation period. Each sub-problem is solved by TS so that the net benefit of the power generation is maximized. For comparison purpose, a non-linear four-reservoir benchmark problem is considered to evaluate the proposed method. Finally, the results are compared with the existing results obtained by GA, PSO, and CA-NLP, showing the efficiency of CA-TS.