| Title |
Estimation of Building Safety Risk in Earthquake Calamity Situation using Linear Regression Algorithm |
| Authors |
황도경(Dokyung Hwang) ; 김다현(Dahyun Kim) ; 김동주(Dongju Kim) |
| DOI |
https://doi.org/10.5573/ieie.2022.59.1.138 |
| Keywords |
Building safety risk; Linear regression; Automatic calculation of building safety risk; Machine learning; Building safety simulator |
| Abstract |
This paper proposes a novel method using linear regression to estimate building safety risk in earthquake calamity situations. The proposed method focused on evaluating the building risk due to ground subsidence and the tilting state of the building among the building risk factors that may occur due to an earthquake. Finally, it calculates the building safety risk by using the estimates for the two states. The conventional method for estimating the risk of a building based on the intensity of the shaking degree as the main symptom of building collapse (inclination and settlement) or the inherent height of the building is analyzed geometrically to estimate the risk state. This paper proposes an automatic risk calculation algorithm based on linear regression to improve the conventional manual risk calculation method. And, to derive the results of the regression algorithm, data are acquired by building a simulation device with a gyro sensor and a GNSS (Global Navigation Satellite System) sensor module attached. Then, the obtained non-linear data are sampled at regular intervals, linearized, and then the state information of the building is estimated through regression. A tested the calculated results for reliability through comparison experiments with actual simulated observations. The risk level is finally determined based on the standards announced by the Ministry of Land, Infrastructure, and Transport of the Republic of Korea using the estimated values. |