M.R. Nejati - Department of Mechanical Engineering Shahid Rajaee Teacher Training University Tehran, Iran
M Hoseinpour Gollo - Department of Mechanical Engineering Shahid Rajaee Teacher Training University Tehran, Iran
M. Tajdari - Department of Mechanic Islamic Azad University Arak, Iran
H. Ghaffarian - School of Computer Engineering Iran University of Science and Technology Tehran, Iran

This paper aimed to propose a novel Fuzzy-based bending angle predictor in laser bending processing. Laser bending is one of the advanced forming methods while most of proposed intelligent bending angle predictors are based on Neural Networks and Genetic Algorithm. This study has presented a pure Fuzzy mechanism in laser bending angle prediction for the first time. Different input data, membership functions and rules have been developed for the Fuzzy predictor. One main note in developing the proposed controller is keeping generality of design. So, it can be adapted in different cases easily. Also, in the case of the lack of knowledge of experts for developing rule base of a Fuzzy controller, a nonlinear regression formula has been used as an alternative approach for developing the rule base. Finally, several experimental tests have been conducted to evaluate performance of the proposed controller. Comparing experimental and predicted results shows that they are in a proper agreement. In some cases, the proposed controller acts even better than the regression models.

laser forming; bending angel; Fuzzy; predictor