So sánh hiệu quả giữa thuật toán hồi quy tuyến tính và phi tuyến tính khi dự báo sóng chấn động nổ mìn trên mỏ than Núi Béo, Quảng Ninh
- Authors: Nguyen Hoang, Bui Xuan Nam, Nguyen Tuan Anh, Nguyen Tuan Thanh
Affiliations:
Hanoi University of Mining and Geology
- *Corresponding:This email address is being protected from spambots. You need JavaScript enabled to view it.
- Received: 15th-Oct-2017
- Revised: 20th-Feb-2018
- Accepted: 10th-Apr-2018
- Online: 30th-Apr-2018
- Section: Mining
Abstract:
Ground vibration ÍPPV) is one of the undesirabie effects induced by blasting operations in open-pit mine, and it can cause damage to surrounding síructures. Therefore, predicting ground vibration is essentialỉy necessary to reduce the environmental effects of blasting. !n this paper, th3 authors used two regression algorithms to predict PPV in Nui Beo open-pit coal mine, inciuding Linear Regression (LR) and Support Vector Machine (SVM). LRỈ represeYìted for linear regression algorithms, whereas SVM represented for non-lìnear regression algorithms. To employ this study, 108° blasíing events were recordẹd with four ^arameters, íncluding ground vibration (PPvJ, elevation betvveen blast sites and moriitor device (H), the capacity of explosive (Q), and the distance betvveen blast sites and monitor device (D). The indicators for evaluating the pertormance of the predictive rnodẹls were used including Root Mean Square Error (RMSE), Coefficient of determination (R2), an9 Mean Absolute Error (MAE). The resuits indicated that LR and SVIVI are suitable for predicting PPV in Nui Beo open-pit coal mine and SVM technique provides highar períormance than LR technique with MAE=1.801, RMSE=2.591, and R2=0.976. In addtion, the results are also the basis of development of other predictive models in blasting operations of open-pit mine in Vietnam.
1. Bhandari. s., Engineering rock blasting operations. A. A. Balkema. 388, 1997: p 388.
2. Hustrulid, W.A., Blasting principies for open pit mining: general design concepts. 1999: Balkema.
3-Hasanipanah, M. và cộng sự. Feasibility of indirect determination of blast induced ground vibratỉon based on support vector machine. Measurement, 2015. 75: p. 289-297.
4. Hasanipanah, M., và cộng sự. Prediction of air-overpressure caused by mine blasting using a new hybrid PSO-SVR model. Engineering with Computers, 2017. 33(1): p. 23-31.
5. Khandelvval, M. và T. Singh, Prediction of blast induced air overpressure in opencast mine. Noise & Vibration Worldwide, 2005. 36(2): p. 7-16.
6. Khandelwai, M. và T. Singh, Prediction of blast-induced ground vibration using artiíicial neural network. International Journal of Rock Mechanics and Mining Sciences, 2009. 46(7): p. 1214-1222.
7. Monjezi, M., H.A. Khoshalan, và A.Y. Varjani, Prediction of ílyrock and backbreak in open pit blasting operation: a neuro-genetic approach. Arabian Ưournal of Geosciences, 2012. 5(3): p. 441-448.
8. Monjezi, M., M. Hasanipanah, và M. Khandelvval, Evaluation and prediction of blast-induced ground vibration at Shur River Dam, Iran,by artiíicial neural network. Neural Computing and Applications, 2013. 22(7-8): p. 1637-1643.
9. Verma, A. và T. Singh, Inteiligent systems for ground vibration measurement: a comparative study. Engineering with Computers, 2011. 27(3): p. 225-233.
10. Longjun, D., và cộng sự Comparisons of random forest and support vector machine for predicting blasting vibration characteristic parameters, Procedia Engineering, 2011. 26: p. 1772-1781. "
11. Hasanipanah, M., và cộng sự Porecasíing blast-induced ground vibration deveioping a CART modeí. Engineering with Computers, 2017. 33(2): p. 307-316.
12. Chandar, K.R., V. Sastry, vả c. Hegde, A Critical Comparison of Regression Models and Artiíicial Neura! Networks to Predict Ground Vibrations. Geotechnịcal and Geological Engineering, 2017. 35(2): p. 573-583. " ’
13. Paradonbeh, R.s. và M. Monjezi, Prediction and minimization of bíast-induced ground vibration using two robust meta-heurìstic algorithms. Engineering with Computers, 2017: p. 1-17.
14. Paradonbeh, R.S., và cộng sự Prediction of ground vibration due to quarry biasting based on gene expression programming: a new model for peak particle velocity predictíon. Internationa! Journal of Environmental Science and Technology, 2016. 13(6): p. 1453-1464.
15. Taheri, K., và cộng sự A hybrid artiíicial bee colony algorithm-artiíicial neural network for íorecasting the blast>produced ground vibration. Engineering with Computers, 2017. 33(3): p. 689700.
16. Sheykhi, H., và cộng sự Porecasting ground vibration due to rock blasting: a hybrid intelligent approach using support vector regression and fuzzy C-means clustering. Engineering with Computers, 2017: p. 1-9.
17. Hasanipanah, M., và cộng sự Estimation of blast-induced ground vibration through a soft computing framework. Engineering Á/ith Computers, 2017: p. 1-9. '
18. VVeisberg, s., Applied linear regressíon. Voỉ. 528. 2005: John VViley & Sons.
19. Seber, G.A. và A.J. Lee, Linear regression analysis. Vol. 936. 2012: John VViley & Sons.
20. Montgomery, D.C., E.A. Peck, và G.G. Vining, Introduction to linear regression analysis. Vol. 821. 2012: John VViley & Sons.
21. Cortes, c. và V. Vapnik, Support-vector netvvorks. Machine learning, 1995. 20(3): p. 273297.
22. Drucker, H., và cộng sự. Support vector regression machines. in Advances in neuraí iníormation Processing systems. 1997.
Other articles
Mining Industry Journal
No. 226 Le Duan Street, Dong Da District, Hanoi City
Tel:(+84-24) 36649158 Fax: (+84-24) 366159
Email:tccongnghiepmo@gmail.com