Adaptive tuning of fuzzy membership functions for non-linear optimization using gradient descent method

Authors:
Prashant S. Vishnupad;Yung C. Shin
Affiliations:
(Currently with Electronic Data Systems) School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USA;School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907, USA
Venue:
Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
Year:
1999

Citing 0
Cited 4

A schema to determine basketball defense strategies using a fuzzy expert system

FS'06 Proceedings of the 7th WSEAS International Conference on Fuzzy Systems
A gradient-descent-based approach for transparent linguistic interface generation in fuzzy models

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
ACO-tuning of a fuzzy controller for the ball and beam problem

MICAI'11 Proceedings of the 10th international conference on Artificial Intelligence: advances in Soft Computing - Volume Part II
A new gravitational search optimization algorithm to solve single and multiobjective optimization problems

Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an adaptive fuzzy tuner for the optimization of non-linear, multi-variable problems. The gradient-descent method is used to adaptively tune the bases of the membership functions used in the fuzzy logic optimization. The performance of the optimization with adaptive tuning is tested in comparison with fuzzy optimization without adaptive tuning of membership functions. It is shown that the adaptive fuzzy optimization provides better performance by converging to the optimal value in lesser time and fewer iterations. A multi-variable, non-linear optimization problem is shown as an illustrative example to demonstrate improved performance. This adaptive tuning scheme has also been incorporated into the Generalized Intelligent Grinding Advisory System (GIGAS II) and results show that even for very complex problems such as manufacturing processes, improved performance can be obtained.