摘要
针对分布式3D打印机(3DPs)在工业物联网(IIoT)中共享、协作、生产全球化的定制产品过程中,3D打印任务(3DPTs)在分布式3D打印机上分配工作量不平衡,以及提交的每个模型的定制属性和实时性等问题,文章提出了用于IIoT中个性化3D打印的实时绿色感知多任务调度架构,给出了一种稳健的在线分配算法,使得每个3D打印任务能够精确地满足用户定义属性,并且平衡了分布式3D打印机之间工作负荷,同时开发了一种基于优先级的自适应实时多任务调度(ARMPS)算法,实时调度每一个3D打印任务,满足3D打印任务的实时性以及动态性要求。在高负载下进行仿真实验,经性能评估测试,表明所提出的算法具有稳健性,调度架构具有鲁棒性和可扩展性。
Aiming at the problems of imbalanced workload distribution of 3D printing tasks(3DPTs)among distributed 3D printers(3DPs)in the process of sharing,collaborating,and producing globalized customized products in the Industrial Internet of Things(IIoT for short),as well as customized attributes and real-time performance of each submitted model,this essay propose a real-time green-aware multi-task scheduling architecture for personalized 3D printing in IIoT,and give a robust online allocation algorithm to solve the problem that each 3D printing task can accurately meet user-defined attributes as well as balance among distributed 3D printers.task scheduling architecture,giving a robust online allocation algorithm to solve the problem that each 3D printing task can accurately satisfy the user-defined attributes as well as balance the workload among distributed 3D printers,and developing a priority-based adaptive real-time multi-task scheduling(ARMPS)algorithm to schedule each 3D printing task in real-time to satisfy the real-time as well as dynamic requirements of 3D printing tasks.Simulation experiments are conducted under high load and performance evaluation tests show that the proposed algorithm is robust and the scheduling architecture is robust and scalable.
作者
赵军富
杜海渊
靳永胜
李建军
ZHAO Junfu;DU Haiyuan;JIN Yongsheng;LI Jianjun(Inner Mongolia University of Science&Technology,Baotou 014010,CHN;Inner Mongolia Mengniu Dairy Group Company Limited,Hohhot 010000,CHN)
出处
《制造技术与机床》
北大核心
2024年第4期188-195,共8页
Manufacturing Technology & Machine Tool
基金
内蒙古自治区直属高校基本科研业务费项目(2023QNJS044)
内蒙古自治区教育科学规划课题(NGJGH2021159)
内蒙古自治区高等学校“青年科技英才支持计划”(NJYT22074)。
关键词
3D打印
工业物联网
任务分配
实时性
多任务调度
3D printing
industrial internet of things
task assignment
real-time
multi-task scheduling
