Concluding Remarks

With the advent of the information age almost every area of human civilization has been touched in some way by the vast computing power that is commonly available. A spillover of the increasing digitization of human activity is the creation of enormous databases and online logging of exabytes of data. Outside of this titanic buildup of data is a vast gap between data storage capabilities and people’s ability to understand and interpret patterns from such large amounts of data. Chemical engineers have not ignored this trend of digitization. For decades, engineers have used data from chemical plants to monitor the state of processes and maintain safe and efficient operation. Nearly all modern processes operate in a highly integrated and automated environment with a distributed control system (DCS) to maintain a consistent product quality and continuously adjust for disturbances entering in the process from the surrounding environment, with the human operator acting as a supervisor.

Data mining and knowledge discovery techniques drawn from computer science literature can help engineers find fault states in historical databases and group them together with little detailed knowledge of the process. This presentation will covers some of the activities under investigation at the Laboratory for Process Systems Engineering at Department of Chemical Engineering at LSU using machine learning tools. Focus will be on recent developments in areas such as: advanced data analytics, intelligent/smart process monitoring and development of novel image-based sensor technologies. Industrial applications will be provided.