Remote Sensing for LNG Processing Plant Equipment Maintenance

Funded by CSIRO Energy SIP Fund
Partners:	R. Gaire, J. Patel
Duration:	2016 – 2017

Information based decision support systems have been used in many areas, including processing plants, for decades. These systems rely on data, and often rules-based analytics, to help make decisions. With the advancement of computers, computer based systems have replaced traditional paper based systems to manage data in databases and to process the data by the application of complex rules. For the maintenance of equipment a variety of decision support systems can be applied, these include: equipment fact sheets; diagnostic checks; maintenance schedules; procedures. These systems are often stored and maintained using computers, typically as spread sheets in software controlled databases. With the emergence of industrial internet of things (IIoT), data that describes the condition of equipment can now be gathered in real time using sensors and sensor network technologies. However, traditional software systems were not designed to manage and analyse the volume of data produced by these sensors. Big data analytics offers a solution for the management and analysis of large data sets.

The use of data from sensors for predictive maintenance is a concept that has gained significant momentum in recent times. This approach can be deployed to train a system using predictive analytics, and to use the trained system to predict future outcomes.  In this project, we studied the technologies related to predictive maintenance with a focus on LNG processing plants, identified opportunities for the development of potentially valuable technologies and solutions and developed a prototype system for monitoring a cooling fan in the context of an LNG processing plant. We also proposed and developed a VR/AR based visualisation system for such systems.