New genetic algorithm for hydropower plants unit commitment optimization

Unit commitment for daily generation scheduling of hydropower plants is a very important issue, and reasonable unit commitment running can bring obvious economic profit, so research of daily generation scheduling has great and far-reaching real-life significance. Through taking the minimum water consumption as objective, the mathematic model is established under daily load task, which is described to two aspects, including state combination and load distribution of units. A new genetic algorithm (NGA) is presented, which adopts dual coding with binary coding and real coding, and performs dual genetic operation with a double crossover and a double mutation for each individual. Realization method of NGA is also designed. In this method, penalty function is used as constraints to reduce the production of non-feasible solution. The result of calculation example shows that NGA is feasible and efficient for daily commitment optimization and its convergence performance is better than GA, with a broad search space and fast convergence and good solution quality.