Vibration and Buckling of Cracked Laminated Composite Beams with Crack Detection Using Fuzzy Logic

Abstract

The laminated composite beams (LCB) witnessed a significant extent of progress in applied engineering applications due to good stiffness to weight, lightweight character and good tailoring ability. However, affected by various static and progressive dynamic loading, crack or crack-like structural defects are inevitable in LCBs. Affected by such load variations, cracks are extremely complex in nature, which raises difficulty for identification and diagnosis of crack-like defects. In this context, the LCBs present a significant and practical research scope under the domain of dynamic and static stability. Apart from the analysis, the identification of damages or structural discontinuities possess a significant research importance to assess the health and functionality of LCBs on a regular basis. Mounting on these significant and potential research aspects, the present investigation is devoted to the structural analysis of LCBs in intact and cracked state by frequency monitoring under free vibration and buckling load variations. Besides, a novel crack detection technique is presented using free vibration and buckling data with Fuzzy Logic Soft Computing tool. The present thesis offers a detailed free vibration and buckling analysis of industry-driven bi-directional woven glass/epoxy and carbon/epoxy LCBs with and without open transverse cracks through experimental and numerical approaches. To this end, a numerical computation for natural frequencies and critical buckling loads is performed on ABAQUS finite element simulation software. A linear beam finite element model is developed on ABAQUS platform and simulated for free vibration and buckling results for non-cracked and cracked LCB. A vibration FFT analyzer is employed for experimental modal analysis and INSTRON 8862 Universal Testing Machine is employed for buckling experiments. It is observed that the numerically simulated results are arrived in line with the experimental outcomes. Based on the developed crack model, the parametric variations are investigated for different parameters. In the present thesis, the free vibration and buckling responses are quantified for various laminated configurations, boundary conditions at supports, span-to-depth ratios and crack signatures. It is shown from the result analysis that the vibration and buckling properties are sensitive to the above-mentioned parameters. Furthermore, significant variations are observed for natural frequencies and buckling loads concerning the crack location and size. In the thesis, the experimental and numerical results concerning each parameter are discussed in detail and conclusions are drawn. During the operational life, the initiation of crack and its existence possesses critical threat to the structural functionality of laminated composites. In this context, the identification of crack location and size is significantly important as it provides a comprehensive knowledge regarding structural state and functionality in advance to avoid such critical threats. To this end, the present research work attempts for crack detection in LCBs using free vibration and buckling responses. Fuzzy Logic tool is employed to train the vibration and buckling data in order to scale the crack location and size. The crack detection using buckling data with Fuzzy Logic has never been attempted before and thus, the present research work is established as a debut attempt. The Fuzzy Logic analysis is performed on MATLAB platform using a Hybrid Mamdani FIS. The simulated vibration and buckling responses for first four modes are fed as input parameters in the developed FIS and the trained data scale the crack signatures as output values. The research attempt for crack detection using vibration and buckling data delivered encouraging results as it expresses similitude with the experimental results.