Design of An Advanced Time Delay Measurement and A Smart Adaptive Unequal Interval Grey Predictor for Real-Time Nonlinear Control Systems

Abstract

This paper is a generation step for developing a novel control methodology based on a variable sampling period (VSP) approach to deal with nonlinear systems containing random delays. The proposed VSP is constructed from an advanced time delay measurement (TDM) method and a novel time delay prediction (TDP) method. The TDM is built to measure real working time of the controlled system, consequently observing a set of actual system delays. Next, the TDP is based on a so-called Smart Adaptive Unequal Interval Grey Model with single-variable first-order—SAUIGM(1,1) to forecast the system delay in the next working step for adjusting the sampling period in order to eliminate bad effects of time delays on the control performance. The SAUIGM(1,1) model was developed from the GM(1,1) model with four significant improvements. It can be easily applied to any practical prediction problem and achieve high prediction accuracy even in case of sparse or largely noisy data. Real-time delay measurements and predictions have been carried out with several examples to verify the proposed TDM and TDP methods. The results indicate that the designed TDM and TDP have strong potential to be applied to the suggested VSP methodology for nonlinear control systems.