Open Access

ARTICLE

Optimum Design for the Magnification Mechanisms Employing Fuzzy Logic–ANFIS

Ngoc Thai Huynh¹, Tien V. T. Nguyen², Quoc Manh Nguyen^3,*

1 Faculty of Automobile Technology, Industrial University of Ho Chi Minh City, Ho Chi Minh City, 70000, Vietnam
2 Faculty of Mechanical Technology, Industrial University of Ho Chi Minh City, Ho Chi Minh City, 70000, Vietnam
3 Faculty of Mechanical Engineering, Hung Yen University of Technology and Education, Hung Yen, 160000, Vietnam

* Corresponding Author: Quoc Manh Nguyen. Email:

Computers, Materials & Continua 2022, 73(3), 5961-5983. https://doi.org/10.32604/cmc.2022.029484

Received 04 March 2022; Accepted 19 May 2022; Issue published 28 July 2022

Abstract

To achieve high work performance for compliant mechanisms of motion scope, continuous work condition, and high frequency, we propose a new hybrid algorithm that could be applied to multi-objective optimum design. In this investigation, we use the tools of finite element analysis (FEA) for a magnification mechanism to find out the effects of design variables on the magnification ratio of the mechanism and then select an optimal mechanism that could meet design requirements. A poly-algorithm including the Grey-Taguchi method, fuzzy logic system, and adaptive neuro-fuzzy inference system (ANFIS) algorithm, was utilized mainly in this study. The FEA outcomes indicated that design variables have significantly affected on magnification ratio of the mechanism and verified by analysis of variance and analysis of the signal to noise of grey relational grade. The results are also predicted by employing the tool of ANFIS in MATLAB. In conclusion, the optimal findings obtained: Its magnification is larger than 40 times in comparison with the initial design, the maximum principal stress is 127.89 MPa, and the first modal shape frequency obtained 397.45 Hz. Moreover, we found that the outcomes obtained deviation error compared with predicted results of displacement, stress, and frequency are 8.76%, 3.6%, and 6.92%, respectively.

---

Keywords

---