Prediction of Earthquake Vulnerability for Low-Rise RC Buildings Using Probabilistic Random Forest
الموضوعات :
Mohammad Khodaparasti
1
,
Ali Alijamaat
2
,
majid pouraminian
3
1 - Department of Computer Engineering,Rasht Branch,Islamic Azad University,Rasht, Iran
2 - Department of Computer Engineering, Rasht Branch, Islamic Azad University, Rasht, Iran
3 - Department of Civil Engineering, Islamic Azad University, Ramsar branch
الکلمات المفتاحية: seismic vulnerability, rapid visual screening, machine learning (ML), random forest (RF), simple Bayesian, reinforced concrete buildings.,
ملخص المقالة :
Assessing the seismic vulnerability of existing buildings is one of the major concerns of governments in the world. Reducing the destructive and catastrophic consequences of earthquakes is necessary and inevitable. So far, various techniques have been presented to evaluate the seismic vulnerability of buildings. One of the fast and effective assessment techniques is the Rapid Visual Screening (RVS) technique with fastly identify high-risk buildings for a more accurate assessment. Among the RVS methods, the Hassan-Sozen PI method is the simplest method to evaluate the seismic vulnerability of low-rise RC buildings. The value of the priority index (PI) is determined from the simple geometric features of the building such as the number of stories, floors area, column area, area of concrete walls and infilled in the main directions of the building. In this article, the data collection have been gathered from Elyasi et al.'s reference such as geometrical information (with geometrical features provided by Hassan-Sozen) and earthquake features (peak ground acceleration and earthquake magnitude) for 658 low-rise RC buildings. The number of considered input features includes seven geometric features and two earthquake features (9 features in total) and the predicted output of Hassan-Sozen priority index. The machine learning technique utilized in this article for prediction seismic vulnerability is a probabilities random forest in which a simple Bayesian method is used to create forest trees. This method has had a slight improvement in accuracy criteria and considerable improvement in accuracy and recall criteria compared to other traditional random forest and ML methods.
[1]. F. F. E. M. A. (US), Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook (FEMA P-154), Applied Technological Council (ATC).
[2]. Harirchian, E.; Aghakouchaki Hosseini, S.E.; Jadhav, K.; Kumari, V.; Rasulzade, S.; Işık, E.; Wasif, M. & Lahmer. T. A. (2021). Review on application of soft computing techniques for the rapid visual safety evaluation and damage classification of existing buildings. Journal of Building Engineering. 43, 102536. https://doi.org/10.1016/j.jobe.2021.102536
[3]. Shaheryar, A.; Abarca, A.; Perrone, D. & Monteiro, R. (2022). Large-scale seismic assessment of RC buildings through rapid visual screening. International Journal of Disaster Risk Reduction. 80, 103219. https://doi.org/10.1016/j.ijdrr.2022.103219
[4]. Hassan, Ahmed F., & Mete A. Sozen. (1997). Seismic vulnerability assessment of low-rise buildings in regions with infrequent earthquakes. ACI structural journal, 94(1). pp31-39
[5]. Elyasi, N.; Kim, E. & Yeum. C.M. (2023). A Machine-Learning-Based Seismic Vulnerability Assessment Approach for Low-Rise RC Buildings. Journal of Earthquake Engineering, 1-17. https://doi.org/10.1080/13632469.2023.2220033
[6]. Chollet F. (2017). Deep learning with Python, Manning Publications, ISBN 9781617296864
[7]. Kunapuli, G. (2023). Ensemble Methods for Machine Learning, Manning Publications, Shelter Island. ISBN 1617297135.
[8]. D. Mienye & Y. Sun, (2022) A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects. IEEE Access, 10, pp. 99129-99149. https://doi.org/10.1109/ACCESS.2022.3207287
[9]. Kuncheva, L. and Whitaker, C., (2003), Measures of diversity in classifier ensembles. Machine Learning, 51, pp. 181-207. https://doi.org/10.1023/A:1022859003006
[10]. Le Cessie S. & Van Houwelingen JC. (1992), Ridge estimators in logistic regression. J R Stat Soc Ser C (Appl Stat),, 41(1), 191–201. https://doi.org/10.2307/2347628
[11]. Pedregosa F.; Varoquaux G.; Gramfort A.; Michel V.; Thirion B.; Grisel O.; Blondel M.; Prettenhofer P.; Weiss R.; Dubourg V.; et al. (2011), Scikit-learn: machine learning in python. J Mach Learn Res., 12, 2825–30.
[12]. G´eron A. (2019), Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: Concepts, tools, and techniques to build intelligent systems, O’Reilly Media.
[13]. Breiman L. (2001). Random forests, Machine Learning, 45, pp5–32. https://doi.org/10.1023/A:1010933404324
[14]. Guo, X. and Hao, P. (2021), Using a Random Forest Model to Predict the Location of Potential Damage on Asphalt Pavement. Appl. Sci., 11, 10396. https://doi.org/10.3390/app112110396
[15]. Belmokre, A., Mihoubi, M.K. & Santillán, D. (2019), Analysis of Dam Behavior by Statistical Models: Application of the Random Forest Approach. KSCE J Civ Eng, 23, 4800–4811. https://doi.org/10.1007/s12205-019-0339-0
[16]. Qinghua, H., Ma, Q., Dang, D.& Xu, J. (2023), Modal Parameters Prediction and Damage Detection of Space Grid Structure under Environmental Effects Using Stacked Ensemble Learning, Structural Control and Health Monitoring, pp1-24. https://doi.org/10.1155/2023/5687265
[17]. Khodaparasti, M.; Alijamaat, A; & Pouraminian, M. (2023). Prediction of the concrete compressive strength using improved random forest algorithm, J Build Rehabil, 8(92). https://doi.org/10.1007/s41024-023-00337-8
[18]. John GH., and Langley P. (1995), Estimating continuous distributions in Bayesian classifiers. In: Proceedings of the Eleventh conference on Uncertainty in artificial intelligence, Morgan Kaufmann Publishers Inc. pp338–345.
[19]. Sarker IH. (2019), A machine learning based robust prediction model for real-life mobile phone data. Internet Things, 5, pp180–93. https://doi.org/10.1016/j.iot.2019.01.007
[20]. Kohavi, R. (1996). Scaling Up the Accuracy of NaïveBayes Classifiers a Decision Tree Hybrid. Second International Conference on Knowledge Discovery and Data Mining, pp. 202-207
