Neural network-based feedforward control of two-stage heat exchange process

An approach to the neural network-based feedforward control of a two-stage heat exchange process is presented. When pressure change occurs in the distillation column, the disturbance causes a variation of material balance that produces undesirable material and heat imbalance. The task is fulfilled first by developing a mathematical model in a traditional way, followed by the implementation of a neural network multilayer feedforward-based controller. The dynamics and unsteady state are investigated