Simulating the behaviour of nanobeams using adaptive neural-fuzzy inference system

Articles in Press

Document Type : Original Article

Authors

Department of Mechanical Engineering, Sirjan University of Technology, Sirjan, Iran

10.48301/kssa.2024.410973.2657

Abstract

In this paper, the analysis of the behavior of cantilever chromium nanobeams has been investigated in the form of nanobeam calculations under static load. for simulate the behavior and calculate the defelection of nanobeams, the adaptive neural-fuzzy inference system(ANFIS), which is a powerful combination of neural network and fuzzy logic, has been used. By using laboratory data, the mentioned system has been trained and tested in three modes. In the first case, the system has been trained with the laboratory results of two forces of 8 and 10.1 nanonewtons along the nanobeam with a thickness of 50 nm. Then, the system has been tested by interpolation with 9.4 nanonewton forces. In the second case, the system has been trained using the experimental 68 nm thick nanobeam for forces of 8 and 11 nanonewtons, and it has been tested with 9.5 and 12.5 nanonewton forces in the form of interpolation and extrapolation. In the third case, with the laboratory results, the nanobeam with a thickness of 83 nm was used under a force of 8 nanonewtons at different points, one at a time, to train the system and to test the results of the system. Percentage error, comparing the ANFIS results with the experimental results, was found to be 2.88%. The results of this research show that it is possible to accurately predict the nanobeam defelection with the adaptive neural-fuzzy inference system without the need to perform more experiments.

Keywords

Main Subjects

Karafan Quarterly Scientific Journal

Articles in Press, Accepted Manuscript
Available Online from 04 February 2024

Files

History

  • Receive Date: 04 October 2023
  • Revise Date: 03 January 2024
  • Accept Date: 30 January 2024

Share

How to cite

Statistics