REAL-TIME STRUCTURAL HEALTH MONITORING OF 3D-PRINTED STRUCTURES WITH EMBEDDED SENSORS
Conference
Laurent, M, Camila, M, Garcia, G et al. (2025). REAL-TIME STRUCTURAL HEALTH MONITORING OF 3D-PRINTED STRUCTURES WITH EMBEDDED SENSORS
. 10.1115/SMASIS2025-165999
Laurent, M, Camila, M, Garcia, G et al. (2025). REAL-TIME STRUCTURAL HEALTH MONITORING OF 3D-PRINTED STRUCTURES WITH EMBEDDED SENSORS
. 10.1115/SMASIS2025-165999
This study explores the use of embedded piezoelectric transducers (PZTs) in 3D-printed polymer structures for real-time structural health monitoring (SHM). Using Fused Filament Fabrication (FFF), sensors were embedded into plates by pausing the print at specific layers, enabling seamless integration without compromising structural fidelity. Multiple infill patterns and densities were used to investigate their influence on wave propagation. The Surface Response to Excitation (SuRE) method was employed to assess the plates under controlled loading conditions. Unlike traditional approaches that use convolutional neural networks and wavelet transforms, this work employs alternative computational techniques, including the Fast Fourier Transform (FFT), the Short-Time Fourier Transform (STFT), and the Empirical Mode Decomposition (EMD). For classification and load estimation, Support Vector Machines (SVM) and Random Forests (RF) were applied. Experimental results demonstrate high accuracy in damage detection and load estimation, with STFT and EMD showing sensitivity to minor applied loads. These findings confirm the feasibility of integrating embedded sensors and advanced signal processing techniques. The proposed two-stage classifier using STFT features achieved >92% accuracy across varying internal geometries.
