Training Feedforward Neural Networks with Gain Constraints
Neural Computation
Dynamic re-optimization of a fed-batch fermentor using adaptive critic designs
IEEE Transactions on Neural Networks
| Hi-index | 0.00 |
There are currently 134 ethanol biorefineries in the United States with a production capacity of nearly 7.2 billion gallons per year, with an additional 6.2 billion gals per year capacity under the construction [1]. Approximately two thirds of these are dry-mill production facilities. Fermentation is a key biorefining process and provides the greatest opportunity for increasing ethanol production. Effective control of the fermentation process is therefore of critical importance to the economic viability of the ethanol production. While this has been the impetus for an increasing interest from researchers in academia and industry, successful control strategies have proven difficult to develop. In this paper we report successful control of ethanol fermentation process in an industrial setting using a parametric nonlinear model predictive control technology. We demonstrate that, using empirical process data and fundamental process knowledge, accurate and numerically efficient models of the fermentation process can be built that enable an optimization-based control of the complex fermentation process. The control strategy is briefly described and representative plots indicating model quality and controller performance are presented.