This article focuses on the development of hybrid machine learning models for predicting the permanent transverse displacement of circular hollow section steel members under impact loads. The increasing demand in structural engineering and related fields for accurate, cost-effective, and real-time solutions has led to the development of hybrid machine learning models, which are capable of combining the advantages of Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs). The effectiveness of the two models developed here is demonstrated through verification and prediction tests conducted using laboratory data collected from a drop-weight impact test. The results show that the hybrid machine learning models can be used to accurately predict the permanent transverse displacement of circular hollow section steel members under impact loads with a reasonably high degree of accuracy.
MDPI is a publisher of peer-reviewed open access journals, established in 1996. It publishes over 350 journals in a wide range of scientific fields, and is driven by its commitment to providing authors with wide dissemination of their research and efficient peer-review by the internationally renowned editorial boards. The articles MDPI publishes are all made immediately available under an open access license, meaning anybody is free to reuse all or part of the contents, provided the original article is properly credited.
The article features research from Dr. Jia Tian, a professor at the University of Wollongong in Sydney, Australia. Dr. Tian is majorly interested in developing computational algorithms and conducting numerical simulations for predicting failure in structures, and his research involved applying the hybrid machine learning models developed here to analysing the drop-weight impact test results. He has published many research papers on engineering and other related topics in various international journals, and is an active member of several international professional societies.
