Site Classification using Feed Forward Backpropagation Artificial Neural Networks.

Abstract

– Strong rock is less affected by the waves propagating during an earthquake. For this reason, structures on strong rocks are less affected by earthquakes. Identifying strong rocks is important for a safe residential area. There are different earthquake codes declaring the characteristics of strong rocks. In this study, site classification was made according to four different earthquake Provisions Nehrp, TBDY, Rm, E code. Feed Forward Backpropagation Artificial Neural Networks was used for site classification. Shear wave velocity (V30), Ground dominant period (To) and H/V ratio were selected as input parameters to this network. Performance analysis was performed to determine which regulation of the Feed Forward Backpropagation Artificial Neural Networks algorithm made the classification more successful. The cross-validation method was used for the analysis. Accuracy, Precision Recall, Kappa, Area under the ROC Curve (AUC) and Root Mean Squared Error (RMS) error values were calculated. As a result, 98% accuracy value was obtained after cross validation in strong rock detection according to E-Code-8 regulation. According to this regulation, all metric values calculated in strong rock detection are higher than other regulations. In addition, hard rock was detected with the least error rate according to this regulation.

Keywords

• Strong Rock
• Nehrp
• TBDY
• Rm
• E-code
• Feed Forward Backpropagation Artificial Neural Networks.

References

1. [1] Kanlı, A. I., Tildy, P., Prónay, Z., Pınar, A., & Hermann, L. (2006). VS 30 mapping and soil classification for seismic site effect evaluation in Dinar region, SW Turkey. Geophysical Journal International, 165(1), 223-235.
2. [2] Foti, S., Parolai, S., Albarello, D., & Picozzi, M. (2011). Application of surface-wave methods for seismic site characterization. Surveys in geophysics, 32, 777-825.
3. [3] Picozzi, M., Strollo, A., Parolai, S., Durukal, E., Özel, O., Karabulut, S., ... & Erdik, M. (2009). Site characterization by seismic noise in Istanbul, Turkey. Soil Dynamics and Earthquake Engineering, 29(3), 469-482.
4. [4] Ulusay, R., & Kuru, T. (2004). 1998 Adana-Ceyhan (Turkey) earthquake and a preliminary microzonation based on liquefaction potential for Ceyhan Town. Natural Hazards, 32, 59-88.
5. [5] Pamuk, E., Özdağ, Ö. C., & Akgün, M. (2019). Soil characterization of Bornova Plain (Izmir, Turkey) and its surroundings using a combined survey of MASW and ReMi methods and Nakamura’s (HVSR) technique. Bulletin of Engineering Geology and the Environment, 78, 3023-3035.
6. [6] Pamuk, E., Özdağ, Ö. C., Tunçel, A., Özyalın, Ş., & Akgün, M. (2018). Local site effects evaluation for Aliağa/İzmir using HVSR (Nakamura technique) and MASW methods. Natural Hazards, 90, 887-899.
7. [7] Salata, S., & Uzelli, T. (2022). Are Soil and Geology Characteristics Considered in Urban Planning? An Empirical Study in Izmir (Türkiye). Urban Science, 7(1), 5.
8. [8] Gülkan, P., Çeken, U., Çolakoğlu, Z., Uğraş, T., Kuru, T., Apak, A., Anderson, J. G., Sucuoğlu, H., Çelebi, M., Akkar, D. S., Yazgan, U. & Denizlioğlu, A. Z. (2007). Enhancement of the national strong-motion network in Turkey. Seismological Research Letters, 78(4), 429-438.

9. [9] Kurtuluş, C. & Bozkurt, A. (2016). Integration of geophysical and geotechnical investigations for Çayırhan town. Journal of Applied Earthscience, 8(2), 15-27.
10. [10] Kurtuluş, C., Sertçelik, İ., Sertçelik, F., Livaoğlu, H., & Saş, C. (2020). Investigation of soil characterization in Hatay Province in Turkey by using Seismic Refraction, Multichannel Analysis of Surface Waves and Microtremor. Earth Sciences Research Journal, 24(4), 473-484.
11. [11] BSSC (Building Seismic Safety Council) (2003). Recommended Provisions for Seismic Regulations for New Buildings and Other Structures and Accompanying Commentary and Maps. FEMA 450, Chapter 3, 17-49.
12. [12] CEN (European Committee for Standardization) (2003). Design of structures for earthquake resistance – part 1: general rules, seismic actions and rules for buildings, EN-1998- 2003, European Committee for Standardization, Brussels
13. [13] TDBY Türkiye (2018). Deprem ve Bina Yönetmeliği, Resmî Gazete, Sayı:30364, 18 Mart.
14. [14] Rodriguez-Marek, A., Bray, J. D. & Abrahamson, N. A. (2001). An empirical geotechnical site response procedure. Earthquake Spectra, 17(1), 65–87.
15. [15] Tildy, P., Hermann, L., & Neducza, B. (2007, September). Problems and Possible Solutions of Geophysics in Eurocode 8 Based Soil Classification. In Near Surface 2007-13th EAGE European Meeting of Environmental and Engineering Geophysics (pp. cp-30). European Association of Geoscientists & Engineers.
16. [16] Paliwal, M., Goswami, H., Ray, A., Bharati, A. K., Rai, R., & Khandelwal, M. (2022). Stability prediction of residual soil and rock slope using artificial neural network. Advances in Civil Engineering, 2022.
17. [17] Mittal, M., Satapathy, S. C., Pal, V., Agarwal, B., Goyal, L. M., & Parwekar, P. (2021). Prediction of coefficient of consolidation in soil using machine learning techniques. Microprocessors and Microsystems, 82, 103830.
18. [18] Jalal, F. E., Xu, Y., Iqbal, M., Javed, M. F., & Jamhiri, B. (2021). Predictive modeling of swell-strength of expansive soils using artificial intelligence approaches: ANN, ANFIS and GEP. Journal of Environmental Management, 289, 112420.
19. [19] Pradeep, T., Bardhan, A., Burman, A., & Samui, P. (2021). Rock strain prediction using deep neural network and hybrid models of anfis and meta-heuristic optimization algorithms. Infrastructures, 6(9), 129.