Machine Learning Method for Spinning Cyber-Physical Production System Subject to Condition Monitoring

摘要

Digitalization encapsulates the importance of machine condition monitoring which is subjected to predictive analytics for realizing significant improvements in the performance and reliability of rotating equipment i.e., spinning. This paper presents a machine learning approach for condition monitoring, based on a regularized deep neural network using automated diagnostics for spinning manufacturing. This article contributes a solution to find disturbances in a running system through real-time data sensing and signal to process via industrial internet of things. Because this controlled sensor network may comprise on different critical components of the same type of machines, therefore back propagation neural network based multi-sensor performance assessment and prediction strategy were developed for our system which worked as intelligent maintenance and diagnostic system. It is completely automatic requiring no manual extraction of handcrafted features.