A novel approach to model determination using the minimum model error estimation

The purpose of this paper is to present an algorithm for the combination of a proven nonlinear system identification technique, the minimum model error estimation algorithm (MME) with an analysis of variance (ANOVA) correlation routine where a forward stepwise procedure is implemented. The analysis of variance approach to model identification is well documented primarily in social science literature but has been sparsely written about for engineering applications. This paper shows a significant improvement in nonlinear model identification when used in conjunction with the MME algorithm.