Real-Time Manufacturing Machine and System Performance Monitoring Using Internet of Things
Real-Time Manufacturing Machine and System Performance Monitoring Using Internet of Things
复制标题
DOI：
10.1109/tase.2017.2784826
复制
发表时间：
2018-10-01
影响因子：
5.6
通讯作者：
Tilbury, Dawn M.
中科院分区：
文献类型：
--
作者：
Saez, Miguel;Maturana, Francisco P.;Tilbury, Dawn M.
介绍了一种利用混合仿真对制造系统进行实时性能评估的框架。监控不同机器的连续和离散变量，以与工厂现场设备同步运行的虚拟环境为参考来分析性能。使用工业物联网解决方案从机器中提取数据。将物理系统的生产率和可靠性与来自混合模拟的数据进行实时比较。该模拟使用离散事件系统来估计系统级别的性能指标，并在机器级别使用连续动态来监控输入和输出变量。模拟输出被用作基于过程不同阶段的实际输出的偏差来检测异常情况的参考。该监控方法在具有机器人和数控机床的全自动化制造系统试验台上实现。机器集成在以太网/IP控制网络上，使用可编程逻辑控制器来协调行动和传输数据。结果表明，能够在可信区间内执行实时监控和捕获性能错误。
This paper introduces a framework to assess the performance of manufacturing systems using hybrid simulation in real time. Continuous and discrete variables of different machines are monitored to analyze performance using a virtual environment running synchronous to plant floor equipment as a reference. Data are extracted from machines using industrial Internet of Things solutions. Productivity and reliability of a physical system are compared in real time with data from a hybrid simulation. The simulation uses discrete-event systems to estimate performance metrics at a system level, and continuous dynamics at a machine level to monitor input and output variables. Simulation outputs are used as a reference to detect abnormal conditions based on deviations of real outputs in different stages of the process. This monitoring method is implemented in a fully automated manufacturing system testbed with robots and CNC machines. Machines are integrated on an Ethernet/IP control network using a programmable logic controller to coordinate actions and transfer data. Results demonstrated the capacity to perform real-time monitoring and capture performance errors within confidence intervals.