Curved beams are lightweight beams. They have applications in many different engineering projects. Therefore, it is very important to study their static and dynamic behavior extensively. The present study aims at presenting efficient approach for health assessment of intact and cracked curved beams based on in-plane and out-of-plane free vibration and buckling analysis. Finite element modelling of intact and cracked curved beams are done using finite element modelling software ANSYS. Experimental investigations of free vibration analysis of intact and cracked curved beam were carried out using Fast Fourier Transform (FFT) spectrum analyser for validation of finite element results from ANSYS. Based on the datasets procured from numerical investigations, Adaptive neuro-fuzzy inference system (ANFIS) models are developed. Four prediction models are developed in the study for estimating in-plane and out-of-plane natural frequency of intact curved beam, in-plane and out-of-plane natural frequency of cracked curved beam, in-plane and out-of-plane buckling load capacity of intact curved beam, and in-plane and out-of-plane buckling load capacity of cracked curved beam. Empirical equations are derived based on ANFIS modelling to develop an inter-relationship between a set of input parameters i.e. (cross-sectional area, radius of curvature, curvature angle, boundary conditions, and mode of vibration viz. in plane or out of plane) and outputs (natural frequency and buckling load capacity) for intact curved beam and (boundary conditions, mode of vibration viz. in plane or out of plane, crack location and crack depth ratios) and outputs (natural frequency and buckling load capacity) for cracked curved beam. Different statistical indices, such as the coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE), is used to evaluate the performance of the input parameters. Sensitivity analysis is used to spot the input parameters that have the greatest influence on the outputs and rank them in importance. The study also developed a detection model for detection of crack parameters (crack location and crack depth ratios) based on free vibration data. Two crack detection methodology viz. Fuzzy logic technique and Multiple Adaptive Neuro-Fuzzy Inference System are showcased and their crack detection ability is compared. 

Keywords: Structural Health Assessment (SHA), Curved Beam, Detection of Cracks, ANFIS, Fuzzy Logic