Integrated data analytics for process systems Lab

Welcome to integrated data analytics for process systems Laboratory

Increasing productivity, safety and efficiency have always been the main goals of industrial plants. The objective of process optimization and control is to reduce the resource wastage and remove bottlenecks while simultaneously satisfying all plant constraints, including operational, economic, and safety constraints.  Obtaining a model that explains the behavior of system over time have significant importance. With the collection of large amount of data in process industries, data-driven modeling especially in the framework of machine learning is gaining popularity these days. However, the data-driven modeling in industrial applications suffers with bottle necks like missing data, delays, outliers, and so on. Further, due to the changes in dynamics of interacting processes over time, and as uncertainties creep in, plant-model mismatch may arise, which must be accounted for, before utilizing such predictive models in process optimization and control.

Our lab primarily focuses on development of robust predictive models for soft sensing, process monitoring and control design, and in development of the data driven optimization schemes for chemical processes.

For further details please visit our research page.