Experimental, Statistical, and Artificial Neural Network Analysis for Steel-Reinforced Elastomeric Bearings Characteristics

Elastomeric bearings play a crucial role in the stability and longevity of structures, particularly in seismic applications. This study investigates the effects of critical parameters, including the number of steel layers, layer thickness, cut rubber thickness, and cross-sectional area, on the shear modulus of steel-reinforced elastomeric bearings. Through a series of experimental tests conducted according to well-established standards, we analyzed the interactions between these parameters using advanced statistical methods and artificial neural networks (ANNs). The findings from this research gain insights into how each parameter contributes to the overall mechanical behavior of elastomeric bearings, which is essential for understanding their performance in dynamic conditions. The results indicate that the number of embedded steel layers has the most significant effect on the mechanical properties of the bearings, followed by layer thickness, which also plays a notable role. Further, a high-fidelity ANN was established that can be deemed as a powerful design tool for engineers to predict the shear modulus of the steel-reinforced elastomeric bearings for a given set of the mentioned parameters. This study serves as a foundational reference for further research into the design and application of elastomeric bearings in bridges and building structures.

کلیدواژه ها:

neoprene rubber ، steel-reinforced elastomeric bearings ، artificial intelligence ، neural networks