A decision‐tree‐based algorithm for identifying the extent of structural damage in braced‐frame buildings

Abstract

Summary Rapid health assessment of essential buildings such as hospitals, fire stations, and large residential complexes is crucial after damaging earthquakes. The use of advanced technologies such as wireless sensors, learning algorithms, and signal processing methods became more attractive in such fast applications due to their higher reliabilities and efficiencies compared to the conventional visual inspection methods. This paper presents a robust post‐earthquake damage detection framework for predicting the extent and location of damage occurrence in the braced‐frame structures after an earthquake. To do so, features derived from acceleration response of the structure were used along with a classification learner to determine the health condition of the structure. Decision tree classifiers are used for the purpose of damage classification where the Bayesian optimization algorithm is implemented to optimize the architecture of the mentioned classifier. A one‐story chevron steel‐braced frame, a three‐story X‐braced steel‐frame, and a five‐story three‐dimensional building are considered to validate the proposed method. The total number of 3774 and 1887 nonlinear response history analyses were respectively performed for 2D and 3D numerical models under scaled SAC motions, using the OpenSees simulation platform. Furthermore, in order to simulate the field condition, a maximum level of 10% white Gaussian noise is added to the output signals. Results obtained from the three case studies show that the proposed framework is robust and reliable in predicting the extent of damage level in the braced‐frame structures in a short time after an earthquake.

Use of Remote Structural Tap Testing Devices Deployed via Ground Vehicle for Health Monitoring of Transportation Infrastructure

“…Receiver (6) Relays information between FS-iA6B transmitter Arduino Uno (7) An input-output device used to help control other components of BRUTUS 1 by utilizing personally developed code.…”

“…Non-contact devices for infrastructure inspections would be of interest if they could inform the classification of rock health automatically. Researchers use sensors and learning algorithms to classify the damage severity in structures [6,7]. This aids in the evaluation of structures in the field, including estimation of the severity of damage and deterioration [8].…”

Seismic damage identification of moment frames based on random forest algorithm and enhanced gray wolf optimization

“…It was confirmed that the method has a robust and reliable performance in quantifying the damage in braced‐frame structures in a short time after the earthquake. [ 15 ]…”

“…It was confirmed that the method has a robust and reliable performance in quantifying the damage in braced-frame structures in a short time after the earthquake. [15] Regarding the different researches such as those reviewed above, there are several problems when encountering by damage identification of real-world structures. An issue is how real structures can be excited to measure their responses for damage identification.…”

A swift neural network-based algorithm for demand estimation in concrete moment-resisting buildings

“…Therefore, estimating these parameters can be significantly efficient in determining the extent of damage in different building types [19]. Machine Learning (ML) is a scientific discipline that investigates the study and development of mathematical algorithms that can construct functional relationship between quantities in terms of known information and rules [20,21]. ML algorithms are widely used for estimating the behavior as well as structural damage of a system under different types of excitations [22].…”