THOMAS J. MCAVOY (mcavoy@isr.umd.edu)

Thomas J. McAvoy received his B.S. in Chemical Engineering from the Brooklyn Polytechnic Institute in 1961. He received M.A. and Ph.D. degrees in Chemical Engineering from Princeton University in 1963 and 1964, respectively.

In 1960 and 1961, Thomas McAvoy worked for the Diamond Alkali Company, Newark, and for the M.W. Kellogg Co., New York, in a Computer Applications Group in 1962. In 1964 he worked for Electronics Associates Inc., Computation Center, Princeton. From 1970 to 1971, Thomas McAvoy served in a Research Apprenticeship at Delft University of Technology in Delft, Netherlands. He taught from 1964 to 1980 at the University of Massachusetts, Chemical Engineering Department. In 1980 he began teaching at the University of Maryland, Chemical Engineering department and received his research faculty appointment to the Systems Research Center in 1985.

Professor McAvoy earned the Allied Foundation Award for Excellence in Undergraduate Education in 1982 and received the Donald P. Eckman Education Award in 1987. He was on the Editorial Board for Instrumentation Technology from 1983 to 1987, Editorial Board I&EC Research from 1989-92, Associate Editor for Automatica from 1991-1993, is currently North America Editor for Automatica, and was a Co-Chairman in 1986 for Chemical Process Control III.

RESEARCH INTERESTS

Dr. McAvoy's interests are in the areas of neural networks, process control, smart sensing, and waste water control.

PUBLICATIONS MOST RELEVANT TO PROPOSED RESEARCH

1. Qin, S., Su, H. and McAvoy, T. J., "Comparison of Four Neural Net Learning Methods for Dynamic System Identification," IEEE Trans. on Neural Networks, 3, 122-130 (1992).

2. Bhat, N. and McAvoy, T. J., "Determining Model Structure for Neural Models By Network Stripping," Computers and Chemical Engineering, 16, 271-281, 1992.

3. Qin, J. and McAvoy, T. J., "Nonlinear PLS Modeling Using Neural Networks," Computers and Chemical Engineering, 16, 379-391, 1992.

4. McAvoy, T. J., Su, H., Wang, N. and He, M., "A Comparison of Neural Networks and Partial Least Squares For Deconvoluting Fluorescence Spectra," Bioengineering and Biotechnology, 40, 53-62, 1992.

5. Su, H., McAvoy, T. and Werbos, P., "Long-Term Predictions of Chemical Processes Using Recurrent Neural Networks: A Parallel Training Approach," I&EC Research, 31, 1338-1352, 1992.

6. Astrom, K. and McAvoy, T. J., "Intelligent Control," J. of Process Control, 2, 115-126, 1992.

7. Su, H. T. and McAvoy, T. J., "Integration of Multilayer Perceptron Networks and Linear Dynamic Models: A Hammerstein Modeling Approach," I&EC Research, 32, 1927-1936, 1993.

8. Qin, S. J. and McAvoy, T. J., "Building Nonlinear FIR Models Via a Neural Net PLS Approach for Long Term Prediction,"accepted by Computers and Chemical Engineering, 1995.

9. Temeng, K., Schnelle, D., Su, H. and McAvoy, T. J., "Neural Model Predictive Control of an Industrial Packed Bed Reactor," presented at Annual AIChE Meeting (Miami Beach, FL, Nov. 1992). J. of Process Control, in press.

10. Dong, D. and McAvoy, T. J., "Nonlinear Principal Component Analysis - Based on Principal Curves and Neural Networks," Computers and Chemical Engineering, in press.