Improved Embedded Structural Health Monitoring via High-Resolution Spectral Analysis Using a MEMCF-based Algorithm

Structural Health Monitoring (SHM) systems based on modal parameters often suffer from sensitivity to environmental and operational variability, limiting their ability in detecting structural alterations. Recently, the Modal Electro-Mechanical Coupling Factor (MEMCF), obtained from piezoelectric sensors operated under different electrical boundary conditions, has been recognized as a more robust damage-sensitive feature. This work investigates the effectiveness of the application of an enhanced spectral resolution algorithm to the modal frequencies computation to cope with the limited memory resources of a low-power embedded sensing platform for damage detection applications. Vibration measurements collected on a laboratory truss subjected to two simulated damage scenarios are used to evaluate the performance of conventional FFT-based frequency estimation and the proposed technique. Four combinations of sampling decimation and window lengths compatible with embedded implementation are examined to assess the minimum spectral resolution required for reliable MEMCF computation. Results show that the proposed algorithm significantly improves the separability between healthy and damaged conditions, outperforming standard FFT processing.