Title: Development of an inverse methodology for crack diagnosis using AI technique

Authors: Amiya Kumar Dash, Dayal R.K. Parhi

Addresses: Department of Mechanical Engineering, Institute of Technical Education and Research, Khandagiri, Bhubaneswar – 751030, Orissa, India. ' Department of Mechanical Engineering, National Institute of Technology, Rourkela – 769008, Orissa, India

Abstract: The influence of cracks on the dynamic behaviour of a cracked cantilever beam with rectangular cross section is discussed in this work. Analytical and experimental investigations are carried out to find the relation between the change in natural frequencies and mode shapes for the cracked and un-cracked beam. Finite element analysis is being performed on the cracked structure to measure the vibration signatures, which is subsequently used in the design of smart system based on fuzzy logic for prediction of crack depths and locations following inverse problem approach. The fuzzy system is developed with relative natural frequencies and mode shapes as input parameters based on the triangular membership functions to calculate the deviation in the vibration parameters for the cracked dynamic structure. Results from experimental analysis are very close to the results predicted by the theoretical, finite element and fuzzy analysis.

Keywords: vibration; multiple cracks; fuzzy logic; mode shapes; natural frequency; cracked beams; cantilever beams; finite element analysis; FEA; crack depths; crack locations; inverse approach.

DOI: 10.1504/IJCMSSE.2011.039570

International Journal of Computational Materials Science and Surface Engineering, 2011 Vol.4 No.2, pp.143 - 167

Available online: 08 Apr 2011 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article