Low-Cost Structural Health Monitoring Using RaspberryShake R4SD Seismic Instrumentation

A cost-effective seismic instrumentation system (LCSIS) was successfully implemented and validated for monitoring the structural health of Peñuelas Tower, the first South American experimental 6-story light-frame timber building. The LCSIS was combined with a relative humidity and temperature measuring system to investigate the influence of ambient conditions on the building's dynamic properties. The LCSIS utilized six accelerographs, specifically RaspberryShake® R4SD, connected through a local network controlled by an auxiliary Raspberry Pi3 (RPi3). The RPi3 served multiple functions, including acting as an NTP server for instrument synchronization, enabling remote control and communication via the internet, and employing internal subroutines to periodically estimate the building's dynamic properties using the FDD method based on velocity data acquired from vertical geophones within the R4SD instruments. To ensure uninterrupted operation, the LCSIS received electrical power from a 55A battery and a battery charger, ensuring functionality during power outages. The LCSIS's validity was confirmed through shake table tests, demonstrating good agreement with conventional measuring systems for signal frequencies below 25–30 Hz and during a low-intensity earthquake with a magnitude of Mw = 4.6. After validating the LCSIS, a BME280 sensor was added to monitor the temperature and relative humidity inside the building continuously. This addition facilitated the implementation of a structural health monitoring system to track ambient-induced variations in the building's three principal natural frequencies. The study revealed that the building's natural frequencies were susceptible to ambient variations, exhibiting daily peak-to-peak variations of 9.5–10.9% and overall variations of 24.7–29.2% over eleven months. These findings indicated that timber structures might be more susceptible to temperature and relative humidity variations than other structural systems. Surprisingly, the timber structure was stiffer under moist conditions, likely due to wood swelling and the resulting tightening of timber assemblies. To accurately compute the estimated natural frequencies, a 36-h state-space model was developed, incorporating recorded temperature and relative humidity data.