Optimal Policy Tracking of a Batch Reactive Distillation by Neural Network-based Model Predictive Control (NNMPC) Strategy

 Abstract—This paper presents the use of neural network-based model predictive control (NNMPC) for handling predefined optimal policy tracking determined by dynamic optimization strategy of a batch reactive distillation column. Multi-layer feedforward neural network model and estimator are developed and used in the model predictive control algorithm. The results show that the NNMPC provides satisfactory control performance for set point tracking problems. The robustness of the NNMPC is investigated with respect to plant/model mismatches and compared to a conventional proportional controller (P). It has been found that the NNMPC provides better control performance than the P controller does in all cases.