Enhanced Harmony Search Algorithm For Sustainable Ecological Operation Of Cascade Hydropower Reservoirs In River Ecosystem

With the merits of superior performance and easy implementation, the harmony search (HS), a famous population-based evolutionary method, has been widely adopted to resolve global optimization problems in practice. However, the standard HS method still suffers from the defects of premature convergence and local stagnation in the complex multireservoir operation problem. Thus, this study develops an enhanced harmony search (EHS) method to improve the HS's search ability and convergence rate, where adaptive parameter adjustment strategy is used to enhance the global search performance of the swarm, while the elite-learning evolutionary mode is used to improve the converge trajectory of the population. To verify its practicability, EHS is applied to solve numerical optimization and multireservoir operation problems. The results show that EHS can produce better results than several existing methods in different cases. For instance, the mean objective of EHS is improved by about 23.9%, 28.7% and 26.8% compared with particle swarm optimization, differential evolution and gravitational search algorithm in 1998-1999 typical runoff case. Hence, an effective optimizer is developed for sustainable ecological operation of cascade hydropower reservoirs in river ecosystem.