Clustering methods - Cause-and-effect diagnosis - Root-cause diagnosis - Nonlinearity analysis - Signal entropy methods - Oscillation detection and diagnosis - Phase coherency analysis - Principal component analysis - Independent component analysis - Spectral analysis - Higher order statistical analysis - Valve stiction modelling and diagnosis
Dynamic signal analysis is our primary specialty. The methods achieve dramatic insights into the way that an operating process is performing, from oil and gas platforms though supply chains to electrical power transmission systems.
An underpinning theme is plant-wide, site-wide and system-wide diagnosis. The aim is to build a global picture using data from local measurements and track down the causes of propagating disturbances.
Sticking valves, a common cause of plant-wide disturbances, can now be easily diagnosed thanks to ideas developed in collaboration with the University of Alberta over the past few years. Now we are seeking to trace the sources of disturbances and inefficiencies that originate in other plant sub-systems particularly in rotating machinery such as compressors and pumps and the motors which drive them.
Selected publications
- Bauer, M., and Thornhill, N.F., 2008, A practical method for identifying the propagation path of plant-wide disturbances, Journal of Process Control, 18, 707-719.
- Bauer , M., Cox, J.W., Caveness, M.H., Downs, J.J., and Thornhill, N.F., 2007, Nearest neighbors methods for root cause analysis of processes with plant-wide disturbances, Industrial Engineering and Chemistry Research, 46, 5977-5984.
- Bauer, M., Cox, J.W., Caveness, M.H., Downs, J.J., and Thornhill, N.F., 2007, Finding the direction of disturbance propagation in a chemical process using transfer entropy, IEEE Transactions on Control System Technology, 15, 12-21.
- Thornhill, N.F. and Horch, A., 2007, Advances and new directions in plant-wide disturbance detection and diagnosis, Control Engineering Practice, 15, 1196-1206.
- Thornhill, N.F., Melbø, H., and Wiik, J., 2006, Multi-dimensional visualization and clustering of historical process data, Industrial Engineering and Chemistry Research, 45, 5971-5985.
- Thornhill, N.F. and Naim, M.M., 2006, An exploratory study to identify rogue seasonality in a steel company's supply network using spectral principal component analysis, European Journal of Operational Research, 172, 146-162.
- Thornhill, N.F., 2005, Finding the source of nonlinearity in a process with plant-wide oscillation, IEEE Transactions on Control System Technology, 13, 434-443.
- Xia, C., Howell, J., and Thornhill, N.F., 2005, Detecting and isolating multiple plant-wide oscillations via spectral independent component analysis, Automatica, 41, 2067-2075.
- Thornhill, N.F., Cox, J.W., and Paulonis, M.A., 2003, Diagnosis of plant-wide oscillation through data-driven analysis and process understanding, Control Engineering Practice, 11, 1481-1490.
- Thornhill, N.F., Huang, B., and Zhang, H., 2003, Detection of multiple oscillations in control loops, Journal of Process Control, 13, 91-100.
- Thornhill, N.F., Shah, S.L., Huang, B., and Vishnubhotla, A., 2002, Spectral principal component analysis of dynamic process data, Control Engineering Practice, 10, 833-846.
Book
- Choudhury, M.A.A.S., Shah, S.L., and Thornhill, N.F., 2008, Diagnosis of Process Nonlinearities and Valve Stiction: Data Driven Approaches, Springer, ISBN: 978-3-540-79223-9.
Contact us
Nina Thornhill, ABB/RAEng Professor of Process Automation
Centre for Process Systems Engineering
Department of Chemical Engineering
Imperial College London
South Kensington Campus, London SW7 2AZ
Tel: +44 (0)20 7594 6622
Email: n.thornhill@imperial.ac.uk