A neural network modeling and control for etherification hybrid process

Abstract

The Etherification Processes are recently improved by the hybrid mix of reaction unit and separation unit. The hybrid processes give a better production but also more complex behaviour. The conventional Proportional Integral Derivative (PID) controllers which are widely used ‘in the industrial chemical processes, are the linear controllers, however, able to control the non-linear and complex processes, but give the slow responses, low performances, limited operating ranges and the performances are not guaranteed in cases of disturbance changes and plant-model mismatches. Recently, the Neural Network control techniques have been successfully applied to these highly non-linear and complex systems due to the recent availability of advanced computer technology. In this research, the neural networks are used as a plant model and controller in a Proportional Integral-Nonlinear Internal Model Control (Pl-NIMC) cascade strategy. The simulation results involve the use of Pl-NIMC cascade control for set point tracking and disturbances rejection in both nominal and plant-model mismatches conditions compared with the conventional PlD cascade control. The Pl-NIMC cascade control strategy was found to be better than the conventional PID cascade control in all cases. These results justify the use of Neural Network control technique in a highly non-linear and complex process which is difficult to control such as this.