Damage Identification in a Cantilever Beam Using Regression and Machine Learning Models

Abstract

A manufacturing fault causes a defect consisting of a crack in the structure. Identification and classification are essential in scientific research because cracks can lead to catastrophic system failure. The purpose of structural fitness tracking is to diagnose and predict structural fitness. A complete crack detection method based on free vibration is widely used to find potential cracks in systems. However, static deflection methods are limited to predicting crack parameters. Therefore, this article uses the static deflection method to determine the crack locations and depth in the cantilever beam. A dead weight was attached to the beam’s free end, and two dial gauges were used. A gauge was attached to the free end of the beam to measure the free-end deflection. Another dial indicator was also installed near the crack to measure the static deflection of the crack. Numerical and experimental analyses were performed on 48 cracked specimens to measure the static deflection at two points. A regression model was developed to calculate the crack parameters, i.e., crack locations and crack depths in beams. To evaluate the reliability of the developed regression model, a machine learning model, i.e., Artificial Neural Network (ANN) and Random Forest (RF), was used for prediction. Regression, ANN, and RF models were developed using numerical and experimental datasets. The crack depth and location results obtained from the regression and machine learning models are consistent with the actual results. The crack parameters were predicted using static two-point deflection as input, and the results were encouraging. Therefore, the static two-point deflection approach may be widely used to detect future cracks in more complex structures.