Xiangjiaba Reservoir Ecological Operation Preliminary Study Based on Genetic Algorithm

Abstract

Traditional reservoir scheduling guarantees efficient utilization of water resources and maximizes hydropower station operation efficiency. However it doesn’t consider the surrounding ecological environment demand. Ecological operation which is based on Traditional reservoir scheduling makes the reservoir achieve long-term, sustainable operation and reduce adverse impacts on the ecological environment. Xiangjiaba reservoir is the ties of the downstream of jinsha river and the Yangtze river which is located in jinsha river basin ecological sensitive area. So the completion of the Xiangjiaba reservoir must result in unhealthy ecological environment impact. With traditional genetic algorithm (GA) encoding complexity and being easy to fall into local convergence limitation, put forward a kind of improved convergence of genetic algorithm, the improved genetic algorithm with real number coding and fitness function which is converted into a nonlinear trigonometric function selection operator. Select the adaptive crossover probability and mutation probability adjust, in order to improve the convergence of the algorithm. RVA method is used to calculate the IHA index upper and lower threshold value and the upper and lower threshold could value as the constraint conditions of ecological scheduling objective function model. The results show that the improved genetic algorithm converge makes the global optimal solution ability stronger and faster; RVA method is suitable for the jinshajiang river and results are more comprehensive and more reasonable.