Welcome to the Adchem '97 Pre-Conference Workshops Page!
This page provides information about Adchem '97's Pre-Conference workshops, to be held just before the actual conference.
| L. Ljung | System Identification This workshop will be held on June 5-6 at the University of Alberta in Edmonton, Alberta. |
|
| Barry Wise, | Process Chemometrics: the application of mathematical and statistical methods for analysis of chemical process data | |
| W. Larimore, N. Kirkendall, D.E. Seborg |
Automated Multivariable System Identification: basic principles and process control applications | |
| T. Marlin, A. Hrymak, F. Forbes, C. Pedersen |
Real Time Optimization |
|
Process Chemometrics: the application of mathematical
and statistical methods for analysis of chemical process data.
Barry M. Wise, Eigenvector Technologies Inc., USA
Email: 73633.2451@compuserve.com
This course is designed to give the student the fundamentals of many commonly used chemometric techniques such as principal components analysis (PCA) and partial least squares (PLS) regression. The couse will also cover non-linear regression methods and pattern recognition. The course will finish with a look at multi-way methods for analysis of multi-dimensional data arrays. The course is aimed at engineers with an interest in Multivariate Statistical Process Control (MSPC), fault detection, soft sensor models, and process modeling for control.
Workshop on System Identification
Lennart Ljung, Sweden
This is a two-day workshop aimed
at teaching the foundations as well as the practice of System Identification.
Participants will be assumed to have some basic knowledge of signals, dynamical
systems and freshman statistics, but need no special background in estimation
and identification.
The workshop covers the basic issues in system identification: estimation
methods, model structures, experiment design and validation techniques.
It explains modern approaches, including prediction error and subspace
methods. It addresses in particular the problem how to decide upon a suitable
model structure.
The focus is user-oriented. The goal of the workshop is that the participants
shall learn how to solve non-trival identification problems in a practical
and efficient way using state-of-the art software. About half of the time
will be devoted to hands-on exercises using MATLAB on mostly real (non-simulated)
data sets.
Automated Multivariable System Identification: basic principles
and process control applications
THIS WORKSHOP HAS BEEN CANCELLED DUE TO AN INSUFFICIENT NUMBER OF REGISTRANTS.
Real-Time Optimization
T. Marlin, A. Hrymak and C. Crowe, McMaster University; Fraser Forbes, University of Alberta; C. Pedersen, Suncor Refinery, Canada
Industrial interest in Real-Time Optimization (RTO) has increased dramatically over the past few years, due in part to the success of model-predictive control methods and the promise of further economic gains. This workshop will provide a discussion of current issues and trends in this evolving process control technology. The workshop will deal with a range of topics including: Gross Error Detection, Data Reconciliation and Parameter Estimations, Online Optimization Algorithms, RTO Results Checking, and a variety of design/implementation issues. The workshop is primarily intended to be relevant to the industrial practitioner and RTO systems developer.
The workshop fees have been set at:
| Workshop Name | Early $US | Later $US (After April 15) |
| System Identification -- L. Ljung (June 5-6) | 850 | 1050 |
| Process Chemometrics -- B. Wise (June 7-8) | 450 | 500 |
| Automated Identification -- Larimore et al. (June 8) | 300 | 350 |
| Real Time Optimisation -- Marlin et al. (June 7-8) | 450 | 500 |
Workshop fees include a set of notes and handouts. Registration for the workshops is on a first-come, first-served basis. Conference participants will receive priority in registering for the workshops. The organizers reserve the right to cancel a workshop should there be an insufficient number of early registrants.
Adchem.97@UAlberta.Ca
http://www.ualberta.ca/dept/chemeng/Adchem/