I’ll be presenting a full-day short course “Nonlinear Regression Modeling” with an emphasis on “Control Applications” as a workshop at the 2018 American Control Conference, 8:30am-5:30pm, Tuesday, June 26, 2018, at the Hilton Milwaukee City Center Hotel, Milwaukee, WI. Participants get a pdf version of course notes and Excel-based software to run various algorithms. The course is aimed at grad students, scientists, engineers, or faculty who wish to pick up application skills.
Kindly share this email with colleagues who might be interested. Thanks.
Visit the conference site for details and to register: http://acc2018.a2c2.org/.
Link to Programs, then Workshops. Until May 10, the workshop fee is $225 for AIChE members and $115 for students or retirees.
Following is a brief description.
Nonlinear Regression Modeling – Emphasizing Control Applications: Techniques and Enabling Design of Experiments
This full-day workshop will be a practical guide for nonlinear regression modeling, with a focus on developing dynamic models for control applications. Although theoretical analysis behind techniques will be revealed, the takeaway will be the participant’s ability to:
- Choose appropriate concepts for defining the regression objective,
- Choose an optimization approach and criteria for convergence,
- Apply both data-based and logical criteria for model validation and model discrimination,
- Design experiments for data generation that support model validation,
- Select an appropriate model design considering both order/complexity and in-use utility, and
- Estimate model uncertainty based on data variability.
The workshop is based on the book, Nonlinear Regression Modeling for Engineering Applications:
Modeling, Model Validation, and Enabling Design of Experiments, by Rhinehart (Wiley, 2016, ISBN 9781118597965).
Participants will receive software for solutions to in-class exercises and implementation techniques in Excel/VBA, a commonly available environment (the techniques could be programmed in any environment). Techniques include the development of FOPDT, SOPDT, ARMA, neural network, and phenomenological models. The course will provide files to implement Leapfrogging as an optimizer, statistical improvement as stopping criteria, bootstrapping for estimating model uncertainty, along with several case-study data sets for revealing course concepts. Participants are invited to bring a laptop with Excel version 2010 or higher for in-class applications. Participants will also receive the workshop presentation material.
Russ Rhinehart is a fellow of ISA, and member of AIChE and the Process Automation Hall of Fame. He has 13-years industrial and 29-years of academic experience.