First Level Pre- and Post-Earthquake Building Seismic Assessment Protocol Based on Dynamic Characteristics Extracted In Situ

Abstract

The present work is concerned with the introduction of a new first level pre- and post-earthquake seismic assessment protocol for buildings that relies on the use of recorded structural response. As earthquakes represent a constant and unpredictable threat for the building stock around the globe, the protocols already in use for assessing the risk should be revised and should also take into account the information hidden in data recorded in the field. Nowadays, data collection does not require expensive equipment and over-qualified personnel. In this direction, the proposed seismic assessment protocol aims to illustrate the ease of widely adopting Structural Health Monitoring (SHM) equipment (e.g., accelerographs), based on the work that has been carried out over the past years on subjects related to earthquake risk estimation. Building taxonomy and damage estimation, like those found in Hazus®–MH and other hazard assessment tools, can be enriched and modified properly to distinguish and classify the very earthquake-prone buildings from the others, and tag them for further assessment and rehabilitation as seismic codes suggest.

The MLDAR Model: Machine Learning-Based Denoising of Structural Response Signals Generated by Ambient Vibration

“…The validation dataset shows that the worst performance achieved was a 91% accuracy for only a few cases, while the average score reached 98% (refer to Figure 25). This confirms the successful accomplishment of the primary goal of this study, which is extracting the eigenfrequencies of SDOF building models from noisy signals, a task previously challenging without extending the sampling time or employing statistical signal manipulation techniques [21].…”

“…Therefore, in order to make use of noisy measurements in various algorithms and methodologies, time-domain dynamic quantities are usually skipped in favor of frequency-domain ones, such as eigenfrequencies. There are various data processing techniques that unveil information hidden by noise, such as the averaging of Fourier Spectra (e.g., [20]) and the recent work by the authors [21]. As said before, in the time domain, little can be done when measurements are already "noisy", and here comes the current work to fill the gap, taking advantage of the power of neural networks and their image detection capabilities.…”

“…In this study, a machine learning-based model is presented that is able to denoise ambient-response recordings collected using instruments with specifications equivalent to those of a low-cost monitoring device (noise density: 22.5 µg/ √ Hz, MEMS type) used in recent SHM studies [21,[26][27][28]. The proposed model is labeled as MLDAR, which stands for Machine Learning-based Denoising of Ambient Response, and is presented graphically in Figure 5.…”