Abstract
The purpose of this study is to develop a neuro process controller for use in a synthesis system of chemical reactions. Chemical reactions are very complex, it is difficult to control the reaction temperature. We developed new type of synthesis systems using a PID controller to control the temperature of the reaction. But, the PID controller was not useful for change of chemical dynamics. Therefore, in this research, a neural network is applied for controlling the temperature of the reaction. First, neural networks controller is constructed to control the temperature of the chemical reaction using an original evaluation function. Second, neural networks estimated suitable parameters for the original evaluation function. The effectiveness is examined through experiments. Finally, using the designed the neuro process controller, it is verified that it can efficiently control chemical reactions
