Machine Learning for Design, Optimization and Assessment of Steel-Concrete Composite Structures: A Review

Abstract

Steel-concrete composite structures (SCCSs) combine the high compressive strength of concrete and tensile strength of steel to achieve optimal structural performance. However, the design of SCCSs is more complex than traditional reinforced concrete (RC) or steel structures due to the steel-concrete composite effects. In recent years, machine learning (ML) has been increasingly applied to SCCSs. However, there have been no related reviews on this topic and this literature gap serves as the motivation for this review. This paper presents the first extensive literature review for ML applications in the design, optimization and assessment of SCCSs. A total of 194 references are collected with most of them are directly related to the ML applications in SCCSs. We discussed ML workflows and models applied for SCCSs, and summarized applications of ML across different SCCS components, including mechanical connectors, steel-concrete interfacial bonding, steel-concrete composite beams, slabs, columns, and walls. The challenges and future research directions are also highlighted. This review provides a valuable reference for researchers and engineers working on the research and development of ML in SCCSs.