Framework and development of fault detection classification using IoT device and cloud environment

초록

While Cyber-physical system (CPS) is considered as a key foundation for cyber manufacturing, many related frameworks and applications have been provided. This research suggests a new and effective CPS architecture for supporting multi-sites and multi-products manufacturing. As target processes, the manufacturing processes for vehicles' High Intensity Discharge (HID) headlight and cable modules are considered. These modules are manufactured with several multi-manufacturing sites consisting of internal manufacturing tasks and intermediate outsourcing processes. In addition, they produce multiple types of HID cable modules with different components. These issues make it difficult to improve the qualities of the overall processes and to control those considering overall manufacturing plants and processes. In order to overcome these limitations, this research provides an Internet of Things (IoT) embedded cloud control architecture. The mixed flow issues are overcome with the cloud control server with the suggested framework. The developed loT device detects several system status and transmits the signals. The data is analyzed for the fault detection classification (FDC) mechanism using deep learning based analytics. Then, the cyber manufacturing based simulation is executed using the provided multi-products queueing network model. The estimated simulation results is used for generating dynamic manufacturing decisions reflecting the real-time changes of the production environment. The suggested framework and its implementations can be used for various industrial processes and applications. (C) 2017 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.

키워드