Time and Frequency Domain Assessment of Low-Power MEMS Accelerometers for Structural Health Monitoring

Abstract

Nowadays, MEMS sensors are widely used in several applications, including SHM. Although they do not outperform professional piezo accelerometers, because of their significantly lower cost and power consumption, they enable extensive, pervasive and battery-less monitoring systems. For these applications, it is critical to assess their performance in typical SHM signal processing tasks and in realistic monitoring scenarios. This paper presents an experimental performance evaluation of representative MEMS devices for SHM applications, providing a guideline and insightful results about the opportunities and capability of these devices in challenging scenarios for such COTS components. Results demonstrate that MEMS-based accelerometers are a feasible solution to replace expensive piezo-based accelerometers. Deploying digital MEMS in Low-Power mode is promising to minimise sensor node energy consumption with savings up to 80% in the best case. Time and frequency domain analysis shows that MEMS can detect modal frequencies, an important parameter for damage detection, with a maximum 1.6% error.