基于多目标遗传算法巨型水库群发电优化调度.pdfVIP

17 6 ( ( ) ; ) Vol. 17 No. 6 2010 11 Earth Science Frontiers ( China University of Geosciences( Beijing) ; Pe ing University) Nov. 2010 1 2 魏加华 , ! 张远东 1 , 100084 2 , 100053 1 2 Wei Jiahua , ! Zhang Yuandong 1State Key Laboratory of Hy droscience and Engineering, Tsinghua University, Beij ing 100084, China 2Water Resources Manage ent Centre, Ministry of Water Resources, Beij ing 100053, China Wei Jiahua, Zhang Yuandong. Optimal operations of multireservoir system using multiobjective genetic algorithms. Earth Sci ence Fr ontiers , 2010, 1 ( 6) :255262 Abstract: In lowcarbon era, hydropow er as clean and renewable energy, energyconservat ion based generation dispatching is an important measure t o reduce energy consumption in China and w ill bring a rare historical op portunity for the hydropower joint dispatching operation. H ow to effectively carry out cascade optimal opera tion of reservoirs, and to fully ma e rat ional use of water resources are urgent needs of cascade hydropow er station management. Based on the analysis of inflow characterist ics of the Qingjiang and T hree Gorges cas cades, a multi objective optimal model for hydropow er dispatching is established. T he tw o objective functions of optimization are to maximize hydropow er generat ion and to m inimize of the t otal discarded water for both in dependent and joint dispatching scenarios. T he multi object ive opt imal model for hydropow er generat ion sys tem is a largescale dynamic nonlinear constrained optimizat ion problem. Mult i objective genetic algorithm ( MOGA) is employed to solve t he model. Data of monthly average runoff from 1951 t o 2002 and daily runoff of typical year are used to longt erm and shortterm optimizat