Auto tuning of PID controller

Abstract

Highly increasing global competition has led to continuous improvisation of process control, optimization and its design which constitute the most important factors of any industry. To analyze the effect of variables at different conditions of process a pilot plant plays a crucial role. This work is directed towards the design, development and implementation of an auto-tuning Proportional Integral Derivative (PID) controller for heat exchanger present in forced circulation evaporator pilot plant and boiler and heat exchanger pilot plant. These laboratory based pilot plants are used to perform rigorous experimentation for validation of theoretical results on actual system. The major part of the process is development of model and implementation of auto tuning algorithm as the tuning parameters have to be updated continuously. Basically in practical implementation of a PID controller and tuning its control parameter there is a possibility that due to human intervention the process is not tuned to obtain optimum control hence auto-tune method serves a masterpiece as it automatically tunes the parameters (which in other case have to be tuned manually) and controls the process to its optimum limits. As a prerequisite, we have considered implementation of both the methods i.e. Ziegler Nichols PID (ZNPID) and Augmented Ziegler Nichols PID (AZNPID) in MATLAB on the generic model of heat exchanger which is validated against results available in the literature and their performance is analyzed for different performance parameters, namely percentage overshoot, rise time and settling time. Ziegler Nichols auto-tune method used for tuning the decentralized PID loop is implemented in MATLAB.