LEARNING OF NEURAL NETWORK TO PREDICT OVERLYING ROCK MASS DISPLACEMENT PARAMETERS BY THE DATA ON JOINTING IN TERMS OF THE ZAPOLYARNY MINE

In terms of the Zapolyarny Mine, creation of a neural network capable to predict angles of overlying rock mass displacements is illustrated. Learning of the neural network was performed by the data on instrumentally measured displacements and based on analytical results of jointing mapping in rock mass surrounding the mine. The initial network was selected as the simplest threelayer neural network composed of inlet, outlet and hidden layers. The inlet layer was intended for primary processing of input signals. For the convenience neurons of the inlet and outlet layers were colored the same way as in the diagrams and graphs plotted. All in all, the input data consisted of 7 indices; therefore, the minimum number of the input neurons was assumed as 7. The outlet neurons were 4 by the same reason. Mathematical implementation of the neural network used the Python language including the Numpy library used for theoretical calculations. Weight factors in the course of the learning process were corrected using feedback communication. The feedback was switched off upon learning completion, and the neural network was used to predict angles of rock movements. After construction, the neural network was tested using the data uninvolved in the learning process. The results prove the neural networks can predict accurately the angels of displacements in rock mass.

 

Acknowledgements: The work was supported by the Russian Science Foundation, Project No. 19-17-00034.


For citation: Sergunin M. P., Eremenko V. A. Learning of neural network to predict overlying rock mass displacement parameters by the data on jointing in terms of the Zapolyarny Mine. MIAB. Mining Inf. Anal. Bull. 2019;(10):106-116. [In Russ]. DOI: 10.25018/0236-1493-2019-10-0-106-116.

Keywords

Jointing, systems of joints, kinematic analysis, displacement, neural language, programming language Python, Dips.

Issue number: 10
Year: 2019
ISBN: 0236-1493
UDK: 550.82 + 622.1
DOI: 10.25018/0236-1493-2019-10-0-106-116
Authors: Sergunin M. P., Eremenko V. A.

About authors: M.P. Sergunin, Head of Department, Center for Geodynamic Safety, Polar Division of PJSC «MMC «Norilsk Nickel», Norilsk, Russia, V.A. Eremenko, Doctor of Technical Sciences, Professor of Russian Academy of Sciences, Director of the Research Center «Application of Geomechanics and Mining of Convergent Technologies», Mining Institute, National University of Science and Technology «MISiS», 119049, Moscow, Russia, e-mail: prof.eremenko@gmail.com. Corresponding author: V.A. Eremenko, e-mail: prof.eremenko@gmail.com.

REFERENCES:

1.     Hong K., Han E., Kang K. Determination of geological strength index of jointed rock mass based on image processing. Journal of Rock Mechanics and Geotechnical Engineering. 2017. No 9. Pp. 702—708.

2.     Zhuravlev E. I. Application of neural network technologies in prediction of geodynamic hazards in mines. Gornyy informatsionno-analiticheskiy byulleten’. 2016, no 2, pp. 66—75. [In Russ].

3.     Moskovskiy I. G., Balaban O. M., Fedorova O. S., Kochetkov A. V. Application of neural networks in interpretation of electromagnetic probing data on vertically jointed media. Naukovedenie. 2015. Vol. 7, no 1. http://naukovedenie.ru/PDF/05TVN115.pdf. DOI: 10.15862/05TVN115. [In Russ].

4.     Thibaut Perol, Michaël Gharbi, Marine Denolle Convolutional neural network for earthquake detection and location. Sci Adv 4 (2), e1700578. (2018) http://advances.sciencemag. org/content/4/2/e1700578, DOI: 10.1126/sciadv.1700578.

5.     Khusainov A. T. Use of artificial neural network instrumentation for operational calculation of oil production. Neftegazovoe delo. 2013. Vol. 11, no 2, pp. 22—25. [In Russ].

6.     Lind Yu. B., Ishbulatova R. Kh., Khashper A. L. Automatic identification of logging programs. Neftegazovoe delo. 2015, no 1, pp. 426—450. DOI: http://dx.doi.org/10.17122/ogbus2015-1-426-450. [In Russ].

7.     Freydin A. M., Tapsiev A. P., Uskov V. A., Nazarova L. A., Zaporozhtsev A. A., Sergunin M. P. Reequipment and technological development in the Zapolyarny Mine. Fiziko-tekhnicheskiye problemy razrabotki poleznykh iskopayemykh. 2007, no 3, pp. 84—94. [In Russ].

8.     Karelin V. N., Marysyuk V. P., Sergunin M. P., Nagovitsin Yu. N., Tapsiev A. P. Introduction of mining systems with self-propelling equipment in the Zapolyarny Mine: Case-study. Fundamental'nye problemy formirovaniya tekhnogennoy geosredy. T. II. Geotekhnologiya [Fundamental Problems of Geo-Environment Formation under Industrial Impact. Vol. II. Geotechnology]. Novosibirsk, Izd. IGD SO RAN, 2010, pp. 47—53.

9.     Bushkov V. K. Use of kinematic stability analysis in substantiation of basic wall design parameters for open pits. Gornyy informatsionno-analiticheskiy byulleten’. 2018, no 10, pp. 30—42. DOI: 10.25018/0236-1493-2018-10-0-30-42. [In Russ].

10.    Pravila okhrany sooruzheniy i prirodnykh ob''ektov ot vrednogo vliyaniya podzemnykh gornykh razrabotok na ugol'nykh mestorozhdeniyakh [Regulations on protection of structures and natural objects from aggravating effect of underground coal mining], Saint-Petersburg, VNIMI, 1998, 291 p. [In Russ].

11.    Dubrovskaya S. A., Ryakhov R. V., Petrishchev V. P. Assessment of potential impact of mining on urban landscape in terms of the Gaya deposit. Ekologiya urbanizirovannykh territoriy. 2016, no 4, pp. 65—71. [In Russ].

12.    Devyatkov S. Yu. Determining conditions of sink formation on ground surface. Strategiya i protsessy osvoeniya georesursov: sbornik nauchnykh trudov, Vyp. 12. [Strategy and processes of development of georesources. Collection of scientific papers, Issue 12], Perm, GI UrO RAN, 2014, pp. 96—98. [In Russ].

13.    Kutepov Yu. Yu., Borger E. B. Numerical modeling of rock mass displacements in terms of geological conditions of the Ruban Mine in Kuzbass. Gornyy informatsionno-analiticheskiy byulleten’. 2017, no 5, pp. 66—75. [In Russ].

14.    Abbas Majdi, Mehdi Amini Flexural toppling failure in rock slopes: from theory to applications. I.J.M.G.E., University of Tehran, Vol. 45, No. 1, December 2011, Pp. 21—32.

15.    Rekomendatsii po parametram sdvizheniya zemnoy poverkhnosti na rudnike «Zapolyarnyy» [Recommendations on ground surface subsidence parameters at the Zapolyarny Mine], Leningrad, VNIMI, 1986. [In Russ].

16.    Otchet po teme 9/60 «Izuchenie protsessa sdvizheniya gornykh porod i proyavleniy gornogo davleniya pod vliyaniem razrabotok na rudnike «Zapolyarnyy» i shakhtakh «TSentral'naya» i «Kayerkan», NGMK im. A.P. Zavenyagina [Report on topic 9/60: Analysis of rock mass displacements and rock pressure events due to mining operations in Zapolayrny, Tsentralnaya and Kaierkan Mines], Norilsk, 1961. [In Russ].

Subscribe for our dispatch