Temperature Control of Continuous Stirred Tank Reactor Based on RTO-ANN-MPC

To solve the problems of nonlinearity, complex reaction mechanism, high energy consumption, etc. of continuous stirred tank reactor (CSTR) in chemical production, this paper combines real-time optimization (RTO) with artificial neural network (ANN) and model predictive control (MPC). The occurrence of endothermic reaction in the reactor is simulated based on the theory of chemical reaction kinetics. The combination of RTO and MPC plays a good role in balancing energy costs and improving control effects. ANN's ability to approximate unknown nonlinear functions solves the problem that RTO and MPC can only be applied to linear models. The results show that this method has a good control effect on temperature.