A Novel Architecture for Condition Based Machinery Health Monitoring on Marine Vessels Using Deep Learning and Edge Computing

Condition based machinery health monitoring on marine vessels involves collecting operational sensor data on the vessel using a robust data acquisition system and determining asset health using anomaly detection analytics. Automation and digitalization of marine vessels involve smart digital technologies such as the Internet of Things (IoT) to collect ships' health data and send it over to a central processing location where this data is analyzed. However, it is difficult to apply this to the shipping industry due to offshore data transmission bandwidth challenges. Deep Learning, a technology that can be used to conduct Machinery Health Monitoring (MHM) holds the key to solve the bandwidth problems. In this paper, we investigate the use of Convolutional Neural Networks (CNN) as a practical solution for deploying Smart Health Monitoring on Marine Vessels using the example of electric induction motors. We show a mechanism to develop data-driven deep learning model that can classify if the motor is in a healthy or faulty condition, and propose an architecture to deploy this model on the Marine vessel in real time on an edge computing hardware. While in operation, sensor data from the motor will be fed into the DL Model, and the resulting predictions will be presented in the Vessel Alarm Monitoring System.