Earticle

초록

영어

This paper addresses the method of multiple cracks detection in moving parts or beams by monitoring the natural frequency and prediction of crack location and depth using Artificial Neural Networks (ANN). Determination of crack properties like depth and location is vital in the fault diagnosis of rotating machine equipments. For the theoretical analysis, Finite Element Method (FEM) is used wherein the natural frequency of beam is calculated whereas the experimentation is performed using Fast Fourier Transform (FFT) analyzer. In experimentation, simply supported beam with single crack and cantilever beam with two cracks are considered. The experimental results are validated with the results of FEM (ANSYSTM) software. This formulation can be extended for various boundary conditions as well as varying cross sectional areas. The database obtained by FEM is used for prediction of crack location and depth using Artificial Neural Network (ANN). To investigate the validity of the proposed method, some predictions by ANN are compared with the results given by FEM. It is found that the method is capable of predicting the crack location and depth for single as well as two cracks. This work may be useful for improving online conditioning and monitoring of machine components and integrity assessment of the structures.

목차

Abstract
 1. Introduction
 2. Analysis of Reduction in Natural Frequencies
 3. Determination of Crack Location
 4. Determination of Crack Size
 5. Experimental Analysis
 6. Finite Element Analysis
 7. Prediction of Crack Properties by Artificial Neural Networks (ANN)
 8. Results and Discussions
 9. Conclusion
 References

키워드

저자

• Prasad Ramchandra Baviskar [ Rajarshi Shahu College of Engineering, Tathwade ]
• Vinod B. Tungikar [ Shri Guru Gobind Singhji Institute of Engg. & Technology ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Advanced Science and Technology

• 간기

  월간

• pISSN

  2005-4238

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605