边-云协同下智能制造单元作业的数字孪生任务调度方法

Digital Twin Task Scheduling Method for Jobs of Intelligent Manufacturing Unit under Edge-cloud Collaboration

智能制造单元作业数字孪生任务的高保真建模和调度优化是智能制造系统实现的关键问题之一。针对该问题,提出一种边-云协同下智能制造单元作业数字孪生任务的调度方法。基于智能制造系统端-边-云架构的虚拟现实交互框架,提出智能制造单元作业与数字孪生任务的映射方法,建立作业数字孪生任务调度的问题模型。考虑智能制造系统虚拟现实交互的快速响应性和偏差问题,提出一种基于端-边-云协同的数字孪生任务混合重调度策略。针对作业数字孪生任务调度的优化目标,设计环境自适应多因子优化遗传算法(Environmental adaptive multi-factor optimization genetic algorithm,EAMO-GA)。试验数据表明,EAMO-GA算法满足结果正确性验证,并且其有效性和收敛性都优于其他算法,可满足大规模、并行式数字孪生任务的调度场景需求。

High fidelity modeling and scheduling optimization of digital twin tasks for jobs in intelligent manufacturing unit is one of the key problems in the implementation of intelligent manufacturing systems.To solve this problem,a scheduling method of digital twin tasks for jobs in intelligent manufacturing unit under edge-cloud cooperation is proposed.Based on the virtual reality interactive framework of the end-edge-cloud architecture of intelligent manufacturing system,a mapping method between jobs of intelligent manufacturing unit and digital twin tasks is proposed,and a scheduling problem model of job digital twin tasks is established.Considering the problem of fast responsiveness and deviation of virtual reality interaction in intelligent manufacturing systems,a hybrid rescheduling strategy of digital twin tasks based on end-edge-cloud collaboration is proposed.Environmental adaptive multi-factor optimization genetic algorithm(EAMO-GA)is designed to optimize the scheduling of job digital twin tasks.The experimental data show that the EAMO-GA meets the correctness verification of the results,and its effectiveness and convergence are better than other algorithms,which can meet the requirements of large-scale and parallel digital twin task scheduling scenario.