Pid-based Temperature Control for Exothermic Chemical Reactor Using Hybrid Ql-ga

Exothermic batch reactions are highly nonlinear and complex processes. During the mixing of the chemical reactants, a sudden and unpredictable heat is released causing exothermic reaction. As a result, the runaway reaction may affect the product quality and pose safety problem to the system. Therefore, it is crucial to control the reactor temperature during the process. Often, good controller requires a precise mathematical model in order to achieve accurate control. However, a precise process kinetic model is not easy to obtain since the process dynamic is changing rapidly. Hence, this paper aims to propose Q-learning (QL) algorithm to observe the process behaviour and to estimate the process parameters through the learning ability. The estimated process parameters are then sent to the genetic algorithm (GA) in order to tune a PID controller through evolutionary process. The results show that the proposed method produces satisfactory results in controlling the reactor temperature.