科技通报
科技通報
과기통보
BULLETIN OF SCIENCE AND TECHNOLOGY
2015年
1期
228-231
,共4页
分区思维%单片机%节能调度
分區思維%單片機%節能調度
분구사유%단편궤%절능조도
partition thinking%single chip microcomputer%efficient scheduling
为了降低单片机系统调度过程的总能耗,提出一种分区思维分类下融合旅行商算法以及遗传算法的节能调度方法。基于分区思维分类方法,将单片机进程调度总能耗,划分成进程切换能耗、进程过渡调整能耗以及进程稳定调度能耗,将单片机进程节能调度问题,转化成单片机多进程调度的能耗优化问题,将单进程调度平稳模态作为节点、进程调度过渡模态作为支路,构建单片机进程调度总能耗的有向图模型,将单片机进程能耗优化过程看成旅行商问题,通过遗传算法的多目标逐层改进以及旅行商算法路径寻优原理,查询不同单片机进程调度下的最佳调度参数和多进程的最佳调度顺序,降低单片机进程调度的总能耗。实验结果说明,所提模型可提高单片机进程调度效率,降低调度能耗。
為瞭降低單片機繫統調度過程的總能耗,提齣一種分區思維分類下融閤旅行商算法以及遺傳算法的節能調度方法。基于分區思維分類方法,將單片機進程調度總能耗,劃分成進程切換能耗、進程過渡調整能耗以及進程穩定調度能耗,將單片機進程節能調度問題,轉化成單片機多進程調度的能耗優化問題,將單進程調度平穩模態作為節點、進程調度過渡模態作為支路,構建單片機進程調度總能耗的有嚮圖模型,將單片機進程能耗優化過程看成旅行商問題,通過遺傳算法的多目標逐層改進以及旅行商算法路徑尋優原理,查詢不同單片機進程調度下的最佳調度參數和多進程的最佳調度順序,降低單片機進程調度的總能耗。實驗結果說明,所提模型可提高單片機進程調度效率,降低調度能耗。
위료강저단편궤계통조도과정적총능모,제출일충분구사유분류하융합여행상산법이급유전산법적절능조도방법。기우분구사유분류방법,장단편궤진정조도총능모,화분성진정절환능모、진정과도조정능모이급진정은정조도능모,장단편궤진정절능조도문제,전화성단편궤다진정조도적능모우화문제,장단진정조도평은모태작위절점、진정조도과도모태작위지로,구건단편궤진정조도총능모적유향도모형,장단편궤진정능모우화과정간성여행상문제,통과유전산법적다목표축층개진이급여행상산법로경심우원리,사순불동단편궤진정조도하적최가조도삼수화다진정적최가조도순서,강저단편궤진정조도적총능모。실험결과설명,소제모형가제고단편궤진정조도효솔,강저조도능모。
In order to lower the total energy consumption of single chip microcomputer system scheduling and put forward a partition thinking under the classification fusion energy-saving scheduling method of traveling salesman algorithm and genetic algorithm. Analyzes the single chip microcomputer scheduling three key part of the total energy consumption, process switching transition energy consumption, adjust the energy consumption of energy consumption and stable operation, mold the MCU to complete the transition process scheduling model of energy consumption, smooth process based on single mode for the node, transition mode for branch, build SCM process scheduling total energy consumption of the directed graph model, a single set of process energy consumption optimization process as a classical traveling salesman problem, through improved step by step a multi-objective genetic algorithm (ga) and traveling salesman algorithm path optimization principle, query the best processing parameters under different SCM process scheduling and the best production order of multiple processes, which lower the total energy consumption in the SCM process scheduling. Experimental results indicate that the proposed model can improve the efficiency of single chip microcomputer process scheduling, reduce energy consumption of scheduling.