Graduate Exam Abstract

Michael Buehner

Ph.D. Preliminary
June 26, 2008, 10:00 AM
Lory Student Center
Two Stage Adaptive Feedforward Predictive Control with Applications to Algal Growth Systems

Abstract: Adaptive and predictive controllers are used to improve the performance of a closed-loop control system; however, they can require on-line stability analysis, which is very time consuming to compute. An adaptive feedforward controller architecture is presented that splits the design into a desired closed reference signal that plant can track followed by the best known control action. A fixed feedback controller is used to correct for the output variations that come from model differences and unmodeled disturbances. The benefits are that the adaptive and predictive controllers are feedforward, which simplifies stability analysis, and the fixed feedback controller operates on small error signals, which can simplify the design. These methods are applied to growing microalgae for biofuel production. In this application, microalgae grow by consuming carbon dioxide and other nutrients in the presence of light. Since there are delivery delays, the controller must predict the amount of carbon dioxide and nutrients to add based on the amount of photons the microalgae will receive.

Adviser: Peter Young
Co-Adviser: Not Applicable
Non-ECE Member: Charles Anderson (Computer Science)
Member 3: Douglas Hittle, Mechanical Engineering
Addional Members: Not Applicable


Program of Study: