Modeling and Control of Highly Nonlinear Complex Process

Many real-world industrial processes exhibit complex nonlinear behavior and cannot be effectively control using conventional linear control methods and algorithms. This paper addresses the issues related to the modeling, identification and control of nonlinear dynamic processes using neural networks. For the measured real data collected and obtained from chemical oil company we created optimal nonlinear NN – MISO model and we design the optimal control algorithm. This NN model has been used to verify different control methods. Results of simulations show performance of NN model and e₹ectiveness of the proposed control approaches.