Decentralized Control of Multi-Input Multi-Output Processes Using Effective Transfer Function Method

Abstract

The design of controllers for multi-input multioutput systems are always complicated and difficult as compared to single-input single-output systems due to the interactions between inputs and outputs. These interactions are addressed in a quantitative manner using Interaction Measures. For ease of designing, the complex multi-input multi-output systems are assumed to be divided into several equivalent single loops and then the controllers are designed for these single loops differently. To simplify the tuning in case of failure tolerant systems, the decentralized controllers are used. The effective transfer function method is used for decentralized control of multi-input multi-output in this paper. This decentralized control method is compared with a centralized control which uses an interaction measure approach based on Hankel Interaction Index Array. Examples using both the methods are employed to show the effectiveness of the decentralized method over the centralized method. The effectiveness is shown by comparing the rise time, settling time and overshoot of the responses obtained from both the methods.