An on-line advisory control system for the lactic acid fermentation process

Fuzzy control of bioprocess in Japan

Process control of bioprocess has been carried out by the judgment of the experts, who are the skilled operators and have lots of experiences for the control of the process. In almost all cases, those experiences are described linguistic IF-THEN rules. Fussy inference is one of the powerful tools to incorporate the linguistic rules to the computer for process control. Fuzzy control are divided into two types; one is the direct fuzzy control of process variables such as sugar feed rate in fed-batch culture and fermentation temperature in batch operation. The other is the indirect control of bioprocess, in which at first the phase recognition is carried out by fuzzy inference and the control strategies constructed in each phase are used for the control of process variables. In Japan, the fuzzy control has already been applied to practical industrial productions, such as pravastatin precursor, vitamin B2, and Japanese sake mashing process. In this review, these industrial approaches of fuzzy control are introduced.

Fuzzy control of bioprocess

Bioprocesses have been operated according to the judgment of experts who are skilled operators and have long experience. In almost all cases, these experiences are described by linguistic IF-THEN rules. Fuzzy inference is one of the powerful tools to incorporate linguistic rules into computational algorithms for application to process control. Fuzzy control is categorized into two types: the direct fuzzy control of process variables such as sugar feed rate in fed-batch culture and broth temperature in batch operation, and the indirect control of bioprocesses in which the phase recognition is first done by fuzzy inference using process variables such as DO, sugar concentration, pH and so on and then the control strategies having been constructed in each phase are used for the process operation. Fuzzy control has already been applied to practical industrial processes, such as production of a pravastatin precursor and vitamin B2, and the Japanese sake mashing process. In this review, these industrial practices of fuzzy control are introduced.

Knowledge-based design and operation of bioprocess systems

Almost twenty years have passed since the first applications of the knowledge-based approach to bioprocess operations were reported. During this period, approaches such as fuzzy logic, artificial neural network modeling, expert systems and genetic algorithms have been extensively studied and successfully used in the design and control of various bioprocesses. The recent development of these approaches in the design and operation of biological processes is summarized and reviewed, especially focusing on the studies reported in biochemical engineering journals.