Authors: Jinyu Li, Shaojun Zhu, Wei Ji, Guo-Qiang Li, Yao Wang, Honghui Qi
Abstract: Fire significantly challenges the integrity and safety of building structures, as it can drastically reduce the strength and stiffness of constructional materials, especially steel, leading to an increased risk of structural failure. However, it is difficult to monitor the structural behavior effectively as traditional measurement techniques fail easily under fire. In response to this challenge, this study aims to develop a high-temperature resistant inclinometer and to advance the methods for acquiring and predicting structural responses during fire incidents. Through comprehensive testing, including failure analysis of a steel beam in a burning furnace and a real fire test on an actual building, this research validates the high-temperature resistance of the newly developed inclinometer. Besides, the effectiveness of indirect displacement measurement methods is also validated—these methods include polynomial fitting and deep learning algorithms. The study demonstrates that the specially designed inclinometer can operate effectively in high-temperature environments for over an hour, providing critical data for monitoring the safety of structures in fire. The displacements obtained from these indirect methods are vital for detecting potential structural collapses caused by fire, significantly contributing to developing an early-warning system for fire-induced collapse of steel structures.