基于多核并行计算的水库群中长期优化调度研究-电气工程专业论文.docxVIP

三 三 峡 大 学 全 日 制 专 业 学 位 硕 士 学 位 论 文 PAGE IV PAGE IV 万方数据 Abstract During the 12th Five-Year Plan period, China has gradually formed the major hydropower bases ,and researching on the reservoir optimal scheduling became a topic of considerable practical significance . Along with expanding the scale of hydropower , calculation for reservoir optimal scheduling in a sharp increase, need to explore new technical methods or means. By detailed analysising of the dynamic programming method (DP), study about the basis of a single reservoir and cascade double reservoirs for long-term optimization scheduling, use the C + + language to design serial program , and parallel by OpenMP on a computer platform within dual-core and quad-core CPU. First , this article combines multi-core technology,which is widely used in personal computers today, with the operating system concepts , focusing on the parallel multicore computing environment , to designed for parallel by application programming interface OpenMP. By actual operating results of simple cycle to procedure test , demonstrating the advantages of parallel computing speed . Secondly, using the knowledge of reservoir optimization scheduling , establishing an annual scheduling period , a monthly schedule time mathematical model of a single reservoir optimal operation for long-term research object , selecting maximum annual generation of the typical year capacity for the objective function , water balance , hydropower output, flood limit water level constraints , and focuses on the dynamic programming method used to design a serial algorithm in this article. Then use OpenMP for loop optimization to parallel algorithm design , which run in different multi-core environments , comparing the time-consuming of two algorithms , analysis of speedup and efficiency of the parallel algorithms . Finally, from the basic upstream and downstream reservoir to start , according to the basin water balance equation to determine the