Crack Detection of Timoshenko Beam Using Frequency and Frequency Response Function

This paper presents a novel approach to detect and estimate cracks in Timoshenko Beams using frequencies and frequency response functions and extreme learning machine. For this purpose, the extreme learning machine used the three first natural frequencies and frequency response functions of beam structure as input which may be noisy or noise free and crack states in beam as output. This data is acquired by the analysis of cracked beams applying the finite element method. To demonstrate the potential of the proposed vibration analysis over existing ones, a validation study has been done. The performance of the presented method has been verified through two numerical examples, namely, a cantilever beam and simply supported beam containing single or multi cracks. Results indicate that the proposed method works well in prediction and estimation of crack and obtained results are accurate. Also, the results show that the presented method is sensitive to the location and severity of crack in spite of the noisy modal data.