Structural health monitoring of South America's first 6-Story experimental light-frame timber-building by using a low-cost RaspberryShake seismic instrumentation. (15th January 2023)

• Record Type: Journal Article
• Authors: Alarcón, Matías; Soto, Pedro; Hernández, Francisco; Guindos, Pablo
• Is Part Of: Engineering structures. Volume 275(2023)Part B
• Publication Date: 2023-01-15
• Subjects: Low-cost seismic Instrumentation -- RaspberryShake -- Structural Health Monitoring -- Ambient Vibration Test -- Light-Frame Timber-Building -- Operational Modal Analysis; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2022.115278
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: Implementation of low-cost seismic instrumentation with RaspberryShake is detailed. SHM of the South America's First 6-Story Experimental Light-Frame Timber-Building. State-Space models can be employed to accurately estimate the natural frequencies. Natural frequencies are susceptible to internal temperature and relative humidity. Natural frequencies can variate 9.5–10.9% daily and 25.2–29.2% during nine months. Abstract: A low-cost seismic instrumentation system (LCSIS) has been implemented and validated for the structural health monitoring of South America's first experimental 6-story light-frame timber building (Peñuelas Tower). Furthermore, coupling the LCSIS with relative humidity and temperature measuring system has allowed for investigating the ambient-induced variations of the building's dynamic properties. The LCSIS has been accomplished using six accelerographs RaspberryShake® R4SD (R4SD), which were interconnected using a local network controlled by an auxiliary Raspberry Pi3 (RPi3). The RPi3 performs the following activities: (i) it works like an NTP server that allows the synchronization between the seismic instruments; (ii) it is connected to the internet to provide remote control and communication; (iii) it has internal subroutines to periodically estimate the dynamic properties of the building by using the FDD method from hourly ambient vibration tests based on the velocity data obtained from the vertical geophones included in R4SD instruments. The LCSIS was electrically supplied using a 55A battery and a battery charger to ensure operability during power outages. The building's natural frequencies and mode shapes were identified from operational modal analyses of five integrated ambient vibration tests, measured by a conventional seismic instrumentation system (Episensors ES-U2), showing consistent dynamic properties. The LCSIS was validated from shake table tests, observing a good agreement for signal frequencies smaller than 25–30 Hz in comparison with conventional measuring systems as well as measuring the response during an earthquake of low-intensity Mw = 4.6. Once the LCSIS was validated, a BME280 sensor was added to continuously monitor the temperature and relative humidity inside the building, which allowed for the implementation of a structural health monitoring system to track the ambient-induced variations of three principal natural frequencies. It was found a high sensitivity of the building's natural frequencies due to ambient variations, exhibiting peak-to-peak daily variations of 9.5–10.9 % and overall variations of 24.7–29.2 % during an eleven-month period, thus evidencing that timber structures could be more susceptible to temperature and relative humidity variations than other structural systems. Unexpectedly, the timber structure was found to be stiffer under moist conditions, which may be attributed to the tightening of timber assemblies due to wood swelling. A 48-hour state-space model was built to accurately compute the estimated natural frequencies as a function of the recorded temperature and relative humidity, showing that this procedure could be used to effectively disaggregate the ambient-induced variations from changes generated by damage or other unknown sources.