An improved genetic algorithm for short-term scheduling of hydroelectric systems

This article proposes a novel approach to solve the short-term optimal generation scheduling of cascaded hydroelectric plants: Quasi-gradient Mixed Genetic Algorithm. The approach adopts new quasi-gradient-embedded three-dimensional real number encoding technique and linear combination mutation operators. The water transport delay time between connected reservoirs is taken into account, and the approach can easily deal with the difficult hydraulic coupling. Numerical results demonstrate that it can improve the accuracy of solutions and increase the speed of convergence, so an efficient method is provided to solve the complex nonlinear constrained optimization problem.