Testing hypotheses used in analysis of control quality

Abstract

Simulation is an important tool for testing and verification of newly designed or modified control algorithms. One of the aims of the simulation verification is a comparison of control quality achieved with new or modified methods with control quality achieved with known methods. For an analysis of control quality, criterions based namely on sum of powers of control errors and sum of powers of control increments are commonly used. These criterions can result only in descriptive attributes of control quality. It means that on the basis of particular values of the criterions it is not possible to identify if the control quality achieved with one algorithm is statistically significantly different from control quality achieved with another algorithm. The aim of this paper is examining of control quality with use of testing hypotheses on existence of statistically significant differences between partial values of the control quality criterions in individual sampling periods. The analysis was performed on a strictly defined significance level 0.001, which is a standardly used value in technical applications. A realization is presented on a simulation of a multivariable predictive control with a modified optimization technique. ©ECMS Mauro Iacono, Francesco Palmieri, Marco Gribaudo, Massimo Ficco (Editors).