PATLAMA KAYNAKLI YER TİTREŞİMİNİN YSA İLE TAHMİNİ VE TAHMİN PERFORMANSI

Bu çalışmada bir taş ocağında gerçekleştirilen patlatma uygulamalarından kaynaklanan yer titreşimleri kaydedilmiş ve bu değerler yapay sinir ağı (YSA) modeli kullanılarak değerlendirilmiş ve tahmin edilmiştir. Ölçümü alınan 28 titreşim verisinin 20 tanesi YSA’nın eğitimi, 4’ü doğrulama ve geriye kalan 4’ü de test için kullanılmıştır. Modelde çıktı parametresi olarak PPV, girdi parametresi olarak ise gecikme başına en fazla patlayıcı miktarı ve ölçekli mesafe kullanılmıştır. Ayrıca MAPE, RMSE ve R2 performans kriterleri, gerçekleşen, YSA ile tahmin edilen ve saha denkleminden elde edilen PPV değerlerinden hesaplanmıştır. Gecikme başına kullanılan en fazla patlayıcı madde miktarı ve ölçekli mesafenin, en yüksek parçacık hızı üzerindeki duyarlılık analizi de belirlenmiştir. Sonuçta, saha denkleminden hesaplanan ve YSA modelinden tahmin edilen titreşim verileri, gerçekleşen titreşim verileri ile karşılaştırıldığında, YSA modeli ile elde edilen değerlerin daha yüksek korelasyona sahip olduğu görülmüştür.

Prediction of Blast-Induced Ground Vibration with ANN and Prediction Performance

In this study, ground vibrations caused by blasting applications in a quarry were recorded and these values were evaluated and estimated by using an artificial neural network (ANN) model. Of the 28 vibration data measured, 20 were used for ANN training, 4 for validation and the remaining 4 for testing. In the model, peak particle velocity (PPV) was used as the output parameter, and the maximum explosive amount per delay and scaled distance were used as input parameters. In addition, MAPE, RMSE and R2 performance criteria were calculated from the realized, predicted by ANN and PPV values obtained from the field equation. The maximum amount of explosives used per delay and the sensitivity analysis of the scaled distance on the highest particle velocity were also determined. As a result, when the vibration data calculated from the field equation and estimated from the ANN model were compared with the realized vibration data, it was seen that the values obtained by the ANN model had a higher correlation.

PDF

___

[1] Özyurt, M.C. (2018). The Investigation of Using Artificial Neural Networks and Game Theory on Underground Mining Method Selection. (Doctoral dissertation, Istanbul University).
[2] Khandelwal, M., & Singh, T. N. (2005). Prediction of blast induced air overpressure in opencast mine. Noise & Vibration Worldwide, 36(2), 7-16.
[3] Liu, Q. L., & Li, X. C. (2014). Modeling and evaluation of the safety control capability of coal mine based on system safety. Journal of cleaner production, 84, 797-802.
[4] Ozer, U., Karadogan, A., Ozyurt, M. C., Sahinoglu, U. K., & Sertabipoglu, Z. (2019). Environmentally sensitive blasting design based on risk analysis by using artificial neural networks. Arabian Journal of Geosciences, 12(2), 60.
[5] Tawadrous, A. (2006). Evaluation of artificial neural networks as a reliable tool in blast design. In proceedings of the annual conference on explosives and blasting technique (Vol. 32, No. 1, p. 71). ISEE; 1999.
[6] Khandelwal, M., & Singh, T. N. (2007). Evaluation of blastinduced ground vibration predictors. Soil Dynamics and Earthquake Engineering, 27(2), 116-125.
[7] Khandelwal, M., & Singh, T. N. (2006). Prediction of blast induced ground vibrations and frequency in opencast mine: a neural network approach. Journal of sound and vibration, 289(4-5), 711-725.
[8] Mohamed, M. T. (2009). Artificial neural network for prediction and control of blasting vibrations in Assiut (Egypt) limestone quarry. International Journal of Rock Mechanics and Mining Sciences, 46(2), 426-431..
[9] Chapra, S.C., Canale, R.P. (2015). Software and Numerical Methods for Engineers with Programming Applications. Literature Publishing, Translators: Hasan Heperkan, Uğur Kesgin, Istanbul, Turkey.
[10] Öztemel, E. (2016), Artificial Neural Networks, (4th Edition). Daisy Publishing, Istanbul, Turkey.
[11] Khandelwal, M., & Singh, T. N. (2009). Prediction of blastinduced ground vibration using artificial neural network. International Journal of Rock Mechanics and Mining Sciences, 46(7), 1214-1222.
[12] Meulenkamp, F., & Grima, M. A. (1999). Application of neural networks for the prediction of the unconfined compressive strength (UCS) from Equotip hardness. International Journal of rock mechanics and mining sciences, 36(1), 29-39.
[13] Lv, C., Xing, Y., Zhang, J., Na, X., Li, Y., Liu, T., ... & Wang, F. Y. (2017). Levenberg–Marquardt backpropagation training of multilayer neural networks for state estimation of a safety-critical cyber-physical system. IEEE Transactions on Industrial Informatics, 14(8), 3436- 3446.
[14] Baghirli, O. (2015). Comparison of Lavenberg Marquardt, Scaled Conjugate Gradient and Bayes Regularization Backpropagation Algorithms for Multistep Ahead Wind Speed Forecasting Using Multilayer Perceptron Feedforward Neural Network. (Master dissertation, Uppsala University)
[15] Siskind, D. E. (1980). Structure response and damage produced by ground vibration from surface mine blasting (Vol. 8507). US Department of the Interior, Bureau of Mines.
[16] Dowding, C. H. (1985). Blast Vibration Monitoring and Control, Prentice-Hall, 297s.
[17] Inan, S., Öztürk, A., & Gürsoy, H. (1993). Stratigraphy of Ulas-Sincan (Sivas) area. Turkish Journal of Earth Sciences, 2, 1-15.
[18] Uyar, G. G., & Aksoy, C. O. (2019). Comparative review and interpretation of the conventional and new methods in blast vibration analyses. Geomechanics and Engineering, 18(5), 545-554.
[19] Yadav, V., & Nath, S. (2017). Forecasting of PM 10 using autoregressive models and exponential smoothing technique. Asian Journal of Water, Environment and Pollution, 14(4), 109-113.
[20] Duvall, W. I. (1963). Vibrations from instantaneous and Millisecond-delayed quarry blasts (No. 6151). US, Department of the Interior, Bureau of Mines.