Title: Data driven model using adaptive fuzzy system

Authors: Nishchal K. Verma, Madasu Hanmandlu

Addresses: Department of Electrical Engineering, Indian Institute of Technology, Delhi, India. ' Department of Electrical Engineering, Indian Institute of Technology, Delhi, India

Abstract: This paper presents the formulation and application of adaptive additive fuzzy adaptive model by using the framework of Gaussian Mixture Model (GMM), which provides the membership functions for the input fuzzy sets. The consequent part of the model is the output function which is derived from the adaptable parameter vector consisting of a weight of a rule, mean and covariance as its elements. These elements are updated using the Expectation and Maximisation (EM) algorithm which is equivalent to Baum-Welch|s backward and forward algorithm for estimating Hidden Markov Model parameters. This resulting model is found to be adaptable depending on the desired input–output behaviour. The model has also been tested on a benchmark problem and the results are found to be better than those obtained from the well known fuzzy models including additive fuzzy models.

Keywords: GMM; Gaussian mixture model; CWM; cluster weighted modelling; additive fuzzy systems; GFM; data driven models; adaptive fuzzy systems.

DOI: 10.1504/IJAAC.2008.022896

International Journal of Automation and Control, 2008 Vol.2 No.4, pp.447 - 458

Published online: 02 Feb 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article