Â鶹ӳ»­Ó°Òô

Abnormal Situation Management and Big Data

Short Description

This project is funded by Lamar Visionary Initiatives and Lamar Distinguished Research Fellow Award.  Process data from wireless sensors and Distributed Control System (DCS) will be used to detect equipment malfunctions early on to facilitate preventive maintenance and to avoid incidents or costly shut-downs. Acoustic and electromagnetic (in addition to density, temperature, and flow) sensors will be explored in the associated experiments. 

Project PI

, Professor, Dan F. Smith Department of Chemical Engineering

 

Full Description

Process data from wireless sensors and Distributed Control System (DCS) will be used to detect equipment malfunctions early on to facilitate preventive maintenance and to avoid incidents or costly shut-downs. Targeted fault detections include HRVOC leaks from heat exchanger networks into cooling towers, sticky or leaky valves, and motor failures.

Advisory/Alert Information will be generated for plant personnel based on sensor and DCS (Distributed Control System) data and analytics. The analytic components include Primary Component Analysis (Latent Variables), Use combination of multi-sensor results, Identify abnormal operations in 2-D plots, Correlation Coefficient, Phase Angle, Frequency Spectrum Analysis, and Model Parameter Identification in addition to fundamental mass and heat balances.  Acoustic and electromagnetic (in addition to density, temperature, and flow) sensors will be explored in the associated experiments.


Research Team/Funding

Co-PIs: Reza Barzegaran, , 
Students: Dan Fernandes, Aniket Khade (Ph.D., Sp. 2016), Napoli Rosario (DE, Sp. 2016)
Funding: Lamar Visionary Initiatives, Lamar Distinguished Research Fellow Award.