一种新的电力系统优化任务调度算法研究
|
摘要
智能电网的调度系统中,存在着巨大的时变调度数据,怎样使用这些数据完成对实时电网信息的提取是本文主要的研究方向。智能电网调度云计算中主要的核心问题就是对电网调度算法的选取,因此,在分析传统方法的基础之上,本文将使用调度中的资源来对电网调度任务进行映射,实现出实数编码。依据编码规则,在进行映射时把所有的大任务分成多个子任务,再依据这些子任务的数量来为染色体的长度进行定义,染色体中对应的基因就是在网络资源中子任务的编号。改进后的遗传算法(IGA)能将种群搜索应用到各个环节中去,改进了传统的调度算法,使得任务完成时间缩短,收敛的速度得到了提高,仿真实验结果表明本文提出的电力系统优化任务调度算法为整个调度系统提高了性能,进一步优化了负载均衡。
There are huge time-varying scheduling data in smart grid scheduling system. How to use these data to extract real-time power grid information is the main research direction in this paper. The main core of smart grid scheduling in cloud computing is the selection of grid scheduling algorithm,therefore,this paper will improve of the shortcomings of existing scheduling algorithms,based on the analysis of the traditional methods,this paper will use the scheduling resources on grid scheduling task mapping,to achieve the real number encoding. According to the coding rule,we divide all the big tasks into many sub tasks while mapping,then we define the length of chromosomes based on the number of these sub tasks.The corresponding genes in chromosomes are the number of neutron tasks in network resources. The improved genetic algorithm( IGA) can be applied to various aspects of population search to improve the traditional scheduling algorithm, the task completion time is shortened, the convergence speed is improved. The simulation results show that the proposed power system optimization scheduling algorithm for the scheduling system to improve performance and optimize the load balance.
There are huge time-varying scheduling data in smart grid scheduling system. How to use these data to extract real-time power grid information is the main research direction in this paper. The main core of smart grid scheduling in cloud computing is the selection of grid scheduling algorithm,therefore,this paper will improve of the shortcomings of existing scheduling algorithms,based on the analysis of the traditional methods,this paper will use the scheduling resources on grid scheduling task mapping,to achieve the real number encoding. According to the coding rule,we divide all the big tasks into many sub tasks while mapping,then we define the length of chromosomes based on the number of these sub tasks.The corresponding genes in chromosomes are the number of neutron tasks in network resources. The improved genetic algorithm( IGA) can be applied to various aspects of population search to improve the traditional scheduling algorithm, the task completion time is shortened, the convergence speed is improved. The simulation results show that the proposed power system optimization scheduling algorithm for the scheduling system to improve performance and optimize the load balance.
引文
[1] Xu Y,Li K,Hu J,et al. A genetic algorithm for task scheduling on heterogeneous computing systems using multiple priority queues[J]. Information Sciences,2014,270(6):255-287.
[2]张松慧,熊锦江.一种针对非平稳网络任务调度防冲突算法研究[J].科技通报,2013(10):143-145.
[3] Wang H C,Woungang I,Yao C W,et al. Energyefficient tasks scheduling algorithm for real-time multiprocessor embedded systems[J]. Journal of Supercomputing,2012,62(2):967-988.
[4] Li K. Energy efficient scheduling of parallel tasks on multiprocessor computers[J]. Journal of Supercomputing,2012,60(2):223-247.
[5] Kaur K, Chhabra A, Singh G. Improved Genetic Algorithm(IGA)for Scheduling Task Graphs in Multiprocessor Systems[J]. Digital Signal Processing,2010.
[6] Wen Y,Xu H,Yang J. A heuristic-based hybrid genetic-variable neighborhood search algorithm for task scheduling in heterogeneous multiprocessor system[J].Information Sciences,2011,181(3):567-581.
[7] Shahul A Z S, Sinnen O. Scheduling task graphs optimally with A*[J]. Journal of Supercomputing,2010,51(3):310-332.
[8] Jiang Y S, Chen W M. Task scheduling for grid computing systems using a genetic algorithm[J]. Journal of Supercomputing,2014,71(4):1-21.
[2]张松慧,熊锦江.一种针对非平稳网络任务调度防冲突算法研究[J].科技通报,2013(10):143-145.
[3] Wang H C,Woungang I,Yao C W,et al. Energyefficient tasks scheduling algorithm for real-time multiprocessor embedded systems[J]. Journal of Supercomputing,2012,62(2):967-988.
[4] Li K. Energy efficient scheduling of parallel tasks on multiprocessor computers[J]. Journal of Supercomputing,2012,60(2):223-247.
[5] Kaur K, Chhabra A, Singh G. Improved Genetic Algorithm(IGA)for Scheduling Task Graphs in Multiprocessor Systems[J]. Digital Signal Processing,2010.
[6] Wen Y,Xu H,Yang J. A heuristic-based hybrid genetic-variable neighborhood search algorithm for task scheduling in heterogeneous multiprocessor system[J].Information Sciences,2011,181(3):567-581.
[7] Shahul A Z S, Sinnen O. Scheduling task graphs optimally with A*[J]. Journal of Supercomputing,2010,51(3):310-332.
[8] Jiang Y S, Chen W M. Task scheduling for grid computing systems using a genetic algorithm[J]. Journal of Supercomputing,2014,71(4):1-21.