Online Adaptive Control for Non Linear Processes Under Influence of External Disturbance

In this paper a novel temperature controller, for non linear processes, under the influence ofexternal disturbance, has been proposed. The control process has been carried out by NeuralNetwork based Proportional, Integral and Derivative (NNPID). In this controller, two experimentshave been conducted with respect to the setpoint changes and load disturbance. The firstexperiment considers the change in setpoint temperature in steps of 10oC from 50oC to 70oC forthree different rates of flow of water. In the second experiment the load disturbance in terms ofaddition of 100ml/min of water at three different time intervals is introduced in the system. It hasbeen shown that, in these situations, the proposed controller adjusts NN weights which areequivalent to PID parameters in both the cases to achieve better control than conventional PID. Inthe proposed controller, an error less than 0.08oC have been achieved under the effect of theload disturbance. Moreover, it is also seen that the present controller gives error less than0.11oC, 0.12oC and 0.12oC, without overshoot for 50oC, 60oC and 70oC, respectively, for allthree rate of flow of water.