Abstract: A mathematical model for the optimal schedule of water injection pumping stations is established, in which the minimum electric power consumption is taken as the objective function, and the restrictions to the displacement and pressure are taken as constraint conditions. In the genetic algorithm dual coding is adopted, the fitness function is adjusted, and random parent-number fitness-weighted cross and many mutation methods are adopted. The simulated annealing algorithm is combined to give the method of determining initial temperature, thus the hybrid genetic algorithm is formed. It can improve the speed of convergence and avoid premature convergence. A processing method of displacement is proposed, so the displacement restrictions are satisfied and the number of infeasible solutions is reduced. Example shows that the algorithm is efficient.