Design of Refined Grey Prediction Controller

The goal of this article is to offer an automatic PID parameter adjustment strategy by using a grey predictor model. The grey predictor model together with a first-order low-pass alpha filter generates effective estimated values of the plant output. The output error is then embedded in a control algorithm that automatically tunes the PID parameters. The resulting PID/Grey controller is shown to outperform the PID controller that is tuned via the standard quarter-decay ratio method. The paper gives a detailed description of the system structure, the design algorithm, and its implementation. Regulation and tracking performance of the proposed PID/grey predictor controller are illustrated by means of computer simulation tests in the Matlab environment. Finally, a temperature regulation example is built by using LabVIEW programming and tested to validate the performance of the proposed controller scheme.