Plant Condition Monitoring by Convolutional Neural Network Integrated into Unified Decision Making System

This paper provides an intelligent plant condition monitoring system using Convolutional Neural Network (CNN) integrated into a decision-making process. CNN unsupervised model can identify faults in complex engineering system, recognize and classify the pattern of failure in an intelligent way that enable plant crew to make quick response for planning maintenance strategy. CNN have been applied to detect machine anomalies in plant and found to be efficient method for fast data processing, identification and classification of normal and abnormal condition of plant rotary components and have shown tremendous results for handling and classification of different faults from massive amount of plant machinery data intrinsic structures irrespective of distortion invariance and scale. CNN overcome dreads using Artificial Intelligence that provides a quick real time condition monitoring and fault detection ability and robustness in decision making with high diagnostic accuracy.