MULTIVARIABLE ON-LINE ADAPTIVE OPTIMIZATION OF A CONTINUOUS CULTURE OF SACCHAROMYCES CEREVISIAE (BAKER'S YEAST, OSCILLATION, TEMPERATURE)
Abstract
The scalar dynamic model identification and optimization algorithm of Bamberger and Isermann was extended to optimize the steady state cellular productivity of a continuous culture of Saccharomyces cerevisiae, or baker's yeast, by manipulating both temperature and dilution rate. The algorithm was successfully implemented on-line, and converged to an optimum steady state in approximately 106 hr. Oscillatory growth was observed during several runs, but the optimization remained stable. When subjected to a step change in pH, the algorithm detected the disturbance and saved the culture from wash out. Adaptive tuning of the optimization gains and the forgetting factor, (lamda) were incorporated into the algorithm. Exponential growth of the optimization gains caused the optimization of oscillate. Optimization with the adaptive (lamda) successfully increased the productivity, but showed sensitivity to the choice of the cell measurement noise variance. Two other studies were performed relating to the optimization. The first was a study of the effect of temperature on the growth of baker's yeast in both batch and continuous cultures. Quantitative results were determined for the effect of temperature on (mu)(,max), K(,s), cell yield, and steady state cell, glucose, and ethanol concentrations. In the second study, oscillatory growth in continuous baker's yeast cultures was studied. The oscillations were characterized, and methods were established for their induction and elimination.
Degree
Ph.D.
Subject Area
Chemical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.