基于Spark的水库群多目标调度粒子群并行化算法

Parallel particle swarm optimization algorithm based on Spark multi-objective optimal scheduling of reservoir group

随着社会经济的发展,在江河修建的水库数量和规模越来越大,水库群优化调度所需资料量庞大、数据关系复杂,应用常规技术进行优化调度存在一定的缺陷,如计算速度慢、存在"维数灾"等。基于此,本文的研究旨在弥补智能算法的缺陷,提高求解模型的速度。本文以黑河流域为例,应用基于Spark框架的水库群多目标调度粒子群并行化算法,并使用Scala语言开发了水库群多目标优化调度软件系统,大大提升了计算效率。研究对水库群多目标优化调度的并行编程发展与应用也有很好的现实意义与应用价值。

With the development of social economy,the number and scale of reservoirs built in rivers are getting ever greater.The optimal scheduling for reservoir group has a large amount of data with the data relationship being complex.The application of conventional technology to optimize scheduling calculation is inefficient,and there is“dimension disaster”,etc.The purpose of this paper is to make up for the shortcomings of the intelligent algorithm and improve the speed of solving the model.Taking the Heihe River basin as an example,multi-objective optimal scheduling of reservoir group for parallel particle swarm optimization algorithm based on Spark is used.The Scala language is used to develop the multi-objective optimal scheduling system of the reservoir group,with the high-speed operation realized.It is of great practical significance and application value for the parallel programming development and application of multi-objective optimization scheduling of reservoir group.