An improved quantum evolution algorithm is presented, which has better global optimization ability and faster convergence speed, because traditional methods have some problems, such as "disaster of dimension" problem, poor convergence performance and so on. The proposed method takes the improved QEA for the outer unit combination and the DP for inner economic load dispatch, the economic operation is solved by the two sub-problems alternating iterative optimization. Meanwhile, the minimum up/down time repair strategy and system reserve capacity repair technique are also used to deal with multiple complex constraints, which effectively improve the convergence speed on the premise of ensuring calculation precision. The proposed method is applied in solving unit commitment problem in Xiluodu Station. Compared with DP and IBPSO methods, the results show that this method is easier and faster, and has better global optimization ability with a strong practical engineering value.