Development of a Random Loading Test Bench for UD E-Glass/Resin Rotating Shafts with Integrated Monitoring of Failure Prediction

• Christian Ivan Basson Discipline of Mechanical Engineering, University of KwaZulu-Natal, Durban, South Africa https://orcid.org/0000-0003-0737-5529
• Glen Bright Discipline of Mechanical Engineering, University of KwaZulu-Natal, Durban, South Africa https://orcid.org/0000-0003-4386-0329
• Jared Padayachee Discipline of Mechanical Engineering, University of KwaZulu-Natal, Durban, South Africa https://orcid.org/0000-0003-0358-5289
• Sarp Adali Discipline of Mechanical Engineering, University of KwaZulu-Natal, Durban, South Africa https://orcid.org/0000-0003-1781-1531

DOI:

https://doi.org/10.7166/36-1-3174

Abstract

Rotating machinery, particularly shafts, is prone to failure owing to cyclic loading, bending stresses, and vibrational oscillations. To enhance their longevity and minimise failures, a predictive maintenance strategy is proposed that integrates Hotelling’s T-squared clustering. Clustering identifies key operational profiles, while embedded sensors gather vibration, temperature, and current data for feature extraction via principal component analysis. The results show that predictive monitoring identifies the remaining useful life of shafts by leveraging data-driven insights, emphasising material-specific characteristics for precise prediction of failure and improved reliability.