能耗感知的云制造服务选择与调度优化方法
Energy consumption aware method for cloud manufacturing service selection and scheduling optimization
彭高贤 1文一凭 1刘建勋 1康国胜 1周旻昊2
作者信息
- 1. 湖南科技大学知识处理与网络化制造湖南省普通高校重点实验室,湖南 湘潭 411201
- 2. 湖南华菱湘潭钢铁有限公司,湖南 湘潭 411201
- 折叠
摘要
云制造服务选择与调度(CMSSS)问题在优化资源配置和满足用户需求方面被广泛关注.然而,大多数现有方法对制造设备的预热过程考虑不足,导致了能源的浪费.为了降低制造能耗并保证服务质量(QoS),建立了 CMSSS的多日标优化模型,通过任务衔接度模型量化制造服务设备的预热能耗,并提出一种能耗感知的云制造服务选择与调度优化方法(ECAM).该方法根据QoS指标为任务选择复合服务,根据制造服务占用情况将子任务调度到空闲时段,并最大化任务衔接度,以降低制造设备的预热能耗.结果表明,在6种评价指标权重下,ECAM比以往的可行调度生成方案(FSGS)具有更好的适应度.在具有预热过程的云制造场景中,ECAM能获得与FSGS基本一致的QoS满意度和更好的能耗经济性.
Abstract
Cloud Manufacturing Service Selection and Scheduling(CMSSS)problem has attracted much attention in optimizing resource allocation and meeting user requirements.However,most existing methods pay insufficient con-sideration to the preheating process of manufacturing equipment,resulted in wasted energy.To reduce manufacturing energy consumption and guarantee Quality of Service(QoS),a multi-objective optimization model for CMSSS was established,the preheating energy consumption of manufacturing service equipment was quantified by a task cohesion degree model,and an Energy Consumption Aware Method(ECAM)for CMSSS optimization was pro-posed.The method selected a composite service for the task according to QoS metrics,and scheduled subtasks to meet the highest cohesion degree in the idle time of the manufacturing service according to the occupation,so as to reduce the preheating energy consumption of the manufacturing equipment.The results showed that ECAM had su-perior fitness to the previous Feasible Schedule Generation Schema(FSGS)under 6 weights evaluation metrics.In cloud manufacturing scenarios with preheating process,ECAM achieved basically the same QoS satisfaction and bet-ter energy economy as FSGS.
关键词
云制造/服务选择与调度/任务衔接度/预热能耗/进化算法Key words
cloud manufacturing/service selection and scheduling/task cohesion/preheating energy/evolutionary al-gorithm引用本文复制引用
基金项目
国家重点研发计划资助项目(2020YFB1707600)
国家自然科学基金资助项目(62177014)
湖南省教育厅资助项目(20B222)
湖南省教育厅资助项目(20C0487)
出版年
2024
NETL
NSTL
万方数据