Process Analysis and Identification with a Real Time Intelligent Bioreactor Supervisory Controller

Bioprocessing can be improved by utilizing all available information about the process. In particular, when knowledge about process dynamics is incomplete or during time of equipment failure, additional information derived from past data trends, expert process operators and the literature can be very useful in process analysis. In this work, an expert system based supervisory control scheme is developed and evaluated. The expert system utilizes on-line real time data as well as archived data, facts and heuristic information to analyze the process. The goal of the analysis is to assess the state of the process and to check for data agreement. In case inconsistency is detected, the expert system will estimate process variables and examine potential process faults. Based on the analysis, acceptable corrective actions and techniques to reconstruct corrupted data will be sought. The behavior and performance of some aspects of a real time implementation of the expert system controller to Clostridium acetobutylicum will be presented and discussed.