fermentation Pilot plant The acquisition of data and the estimation of state parameters on commercial scales will undoubtedly become increasingly significant Unfortunately, the advanced control involving adaptive and optimize controls have not yet been sufficiently investigated in either the pilot or industrial scale Adaptive control is of great importance for self-optimization of fermentation processes, even on a commercial scale, because in ordinary fermentation the process includes several variables regarding culture condi tions and raw materials. We are sometimes faced with difficulties in the mathematical modelling of fermentation processes because of the complex reaction kinetics involving cellular metabolism. The knowledge-based controls using fuzzy theory or neural networks have been found very uset for what we call the"black box processes. Although the complexity of the process and the number of control parameters make control problems in fermentation very difficult to solve, the solution of adaptive optimization strategies is worthwhile and can contribute greatly to total profits. In order to establish such investigations, many fermentation corporations have been building pilot fermentation systems that consist of highly instrumented fermenters coupled to a distributed hierarchical computer network for on-and off-line data acquisition, data analysis, control and modelling. An example of the hierarchical computer system that is shown in Fig. 3 has become as common in the installation of large fermentation plants as it is elsewhere in the chemical industry. Figure 4 shows the details of a computer communi- cation network and hardware As seen in Fig 3, the system is mainly divided into three different functional levels. The first level has the YEWPACK package instrumenta tion systems(Yokogawa Electric Corporation, Tokyo), which may consist of an operators console (UOPC or UOPS)and several field control units (UFCU or UFCH) which are used mainly for on-line measurement, alarm, sequence control, and various types of proportional-integral-derivative(PID) controls. Each of the field control units interfaces directly with input/output signals from the instruments of fermenters via program controlle conditioners. In the second level, YEWMAC line computer systems (Yokogawa Electric Corporation, Tokyo) are dedicated to the acquisiti storage, and analysis of data as well as to documentation, graphics, optimi zation, and advanced control. a line computer and several line controllers constitute a YEWMAC. The line controller also governs the local area network formed with some lower level process computers using the bsc multipoint system. On the third level, a mainframe computer is reserved for modelling, development of advanced control, and the building of a data baseFermentation Pilot Plant 9 The acquisition of data and the estimation of state parameters on commercial scales will undoubtedly become increasingly significant. Unfortunately, the advanced control involving adaptive and optimized controls have not yet been sufficiently investigated in either the pilot or industrial scale. Adaptive control is of great importance for self-optimization of fermentation processes, even on a commercial scale, because in ordinary fermentation the process includes several variables regarding culture condi￾tions and raw materials. We are sometimes faced with difficulties in the mathematical modelling of fermentation processes because of the complex reaction kinetics involving cellular metabolism. The knowledge-based controls using fuzzy theory or neural networks have been found very useful for what we call the “black box” processes. Although the complexity of the process and the number of control parameters make control problems in fermentation very difficult to solve, the solution of adaptive optimization strategies is worthwhile and can contribute greatly to total profits. In order to establish such investigations, many fermentation corporations have been building pilot fermentation systems that consist of highly instrumented fermenters coupled to a distributed hierarchical computer network for on-and off-line data acquisition, data analysis, control and modelling. An example of the hierarchical computer system that is shown in Fig. 3 has become as common in the installation of large fermentation plants as it is elsewhere in the chemical industry. Figure 4 shows the details of a computer communi￾cation network and hardware. As seen in Fig. 3, the system is mainly divided into three different functional levels. The first level has the YEWPACK package instrumenta￾tion systems (Yokogawa Electric Corporation, Tokyo), which may consist of an operator’s console (UOPC or UOPS) and several field control units (UFCU or UFCH) which are used mainly for on-line measurement, alarm, sequence control, and various types of proportional-integral-derivative (PID) controls. Each of the field control units interfaces directly with input/output signals from the instruments of fermenters via program controllers and signal conditioners. In the second level, YEWMAC line computer systems (Yokogawa Electric Corporation, Tokyo) are dedicated to the acquisition, storage, and analysis of data as well as to documentation, graphics, optimi￾zation, and advanced control. A line computer and several line controllers constitute a YEWMAC. The line controller also governs the local area network formed with some lower level process computers using the BSC multipoint system. On the third level, a mainframe computer is reserved for modelling, development of advanced control, and the building of a data base