Structural health monitoring of beam-column connections: a novel method combining empirical mode decomposition and the Newton-Raphson method

Purpose This study proposes an integrated method for structural health monitoring of beam-column connections in steel moment frames, aiming to enhance damage detection accuracy in noisy environments. Traditional approaches often fall short in the presence of measurement uncertainties and semi-rigid joint behavior. By incorporating empirical mode decomposition (EMD) with finite element model updating, this research addresses these challenges and provides a robust solution. Design/methodology/approach The proposed framework involves three main steps. First, sensor-recorded acceleration responses are decomposed into intrinsic mode functions (IMFs) via EMD. Second, a thresholding technique is applied to these decompositions to mitigate noise and reconstruct denoised signals. Finally, a Newton-Raphson-based model updating procedure is used to localize and quantify damage in beam-column connections. Numerical evaluations are performed on two-dimensional steel moment frames subjected to various noise levels and damage scenarios. Findings The results indicate that EMD-based noise attenuation significantly improves damage detection outcomes. Compared to a standard Newton-Raphson approach using unfiltered data, the integrated method demonstrates higher accuracy in localizing damaged connections and yields lower mean squared errors. Additionally, it effectively identifies damage in both joints and structural members, showcasing its robustness and applicability in practical scenarios. Originality/value By combining advanced signal processing with an iterative model updating algorithm, this study presents a novel and efficient structural health monitoring (SHM) methodology specifically designed for semi-rigid steel frames. The synergy between EMD-based denoising and the Newton-Raphson procedure offers enhanced damage detection capabilities, adding tangible value for researchers and practitioners in structural health monitoring.