Application of Cloud Model in Rock Burst Prediction and Performance Comparison with Three Machine Learning Algorithms

Rock burst is a common disaster in deep underground rock mass engineering excavation. In this paper, a cloud model (CM) is applied to classify and assess rock bursts. Some main factors that influence rock bursts include the uniaxial compressive strength (<inline-formula> <tex-math notation="LaTeX">$\sigma _{\mathrm {\mathbf {c}}}$ </tex-math></inline-formula>), the tensile strength (<inline-formula> <tex-math notation="LaTeX">$\sigma _{\mathrm {\mathbf {t}}}$ </tex-math></inline-formula>), the tangential stress (<inline-formula> <tex-math notation="LaTeX">$\sigma _{\mathrm {\boldsymbol{\theta }}}$ </tex-math></inline-formula>), the rock brittleness coefficient (<inline-formula> <tex-math notation="LaTeX">$\sigma _{\mathrm {\mathbf {c}}}/\sigma _{\mathrm {\mathbf {t}}}$ </tex-math></inline-formula>), the stress coefficient (<inline-formula> <tex-math notation="LaTeX">$\sigma _{\mathrm {\boldsymbol{\theta }}}/\sigma _{\mathrm {\mathbf {c}}}$ </tex-math></inline-formula>), and the elastic energy index (<inline-formula> <tex-math notation="LaTeX">$W_{\mathrm {\mathbf {et}}}$ </tex-math></inline-formula>), which are chosen to establish the evaluation index system. The weights of these indicators are obtained by the rough set method based on 246 sets of domestic and foreign rock burst samples. The 246 samples are classified by normalizing the data and establishing an RS-CM. The 10-fold cross validation was used to obtain higher generalization ability of models. The classification results of the RS-CM are compared with those of the Bayes, KNN, and RF methods. The results show that the RS-CM exhibits higher values of accuracy, Kappa, and three within-class classification metrics (recall, precision, and the F-measure) than the Bayes, KNN, and RF methods. Hence, the RS-CM, which is characterized by high discriminatory ability and simplicity, is a reasonable and appropriate approach to rock burst classification and prediction. Finally, the sensitivity of six indexes was investigated to take scientific and reasonable measures to prevent or reduce the occurrence of rock bursts.

[1]  C Fairhurst,et al.  ROCKBURSTS: PREDICTION AND CONTROL. PAPERS PRESENTED AT A SYMPOSIUM ORGANIZED BY THE INSTITUTION OF MINING AND METALLURGY IN ASSOCIATION WITH THE INSTITUTION OF MINING ENGINEERS, AND HELD IN LONDON 20 OCTOBER, 1983 , 1983 .

[2]  Zhigang Luo,et al.  Semi-Supervised Projective Non-Negative Matrix Factorization for Cancer Classification , 2015, PloS one.

[3]  Wei Wang,et al.  Innovative product design based on customer requirement weight calculation model , 2010, Int. J. Autom. Comput..

[4]  J. Koenderink.  Q… , 2014, Les noms officiels des communes de Wallonie, de Bruxelles-Capitale et de la communaute germanophone.

[5]  C. Tang,et al.  Rock failure mechanisms : explained and illustrated , 2010 .

[6]  Yong Wang,et al.  A rough set approach for determining weights of decision makers in group decision making , 2017, PloS one.

[7]  Morteza Zahedi,et al.  Improving Text Classification Performance Using PCA and Recall-Precision Criteria , 2013 .

[8]  K. Gwet,et al.  A comparison of Cohen’s Kappa and Gwet’s AC1 when calculating inter-rater reliability coefficients: a study conducted with personality disorder samples , 2013, BMC Medical Research Methodology.

[9]  Lg Tham,et al.  Method of Fuzzy Comprehensive Evaluations for Rockburst Prediction (in Chinese) , 1998 .

[10]  G. G. Stokes.  "J." , 1890, The New Yale Book of Quotations.

[11]  A. Kidybiński,et al.  Bursting liability indices of coal , 1981 .

[12]  Salvatore Greco,et al.  Discovering Reservoir Operating Rules by a Rough Set Approach , 2006 .

[13]  Chen Shou-ru,et al.  APPLICATION OF UNASCERTAINED MEASUREMENT MODEL TO PREDICTION OF CLASSIFICATION OF ROCKBURST INTENSITY , 2010 .

[14]  S. P. Singh,et al.  Classification of mine workings according to their rockburst proneness , 1989 .

[15]  Guohui Zhang,et al.  Development of new variance reduction methods based on weight window technique in RMC code , 2016 .

[16]  S. Batugin,et al.  State of stress in the upper part of the Earth's crust based on direct measurements in mines and on tectonophysical and seismological studies , 1972 .

[17]  Deyi Li,et al.  Artificial Intelligence with Uncertainty , 2004, CIT.

[18]  Jian-Fu Shao,et al.  Comprehensive Stability Evaluation of Rock Slope Using the Cloud Model-Based Approach , 2014, Rock Mechanics and Rock Engineering.

[19]  W. Marsden.  I and J , 2012 .

[20]  Qiu Dao-hong,et al.  Application of RBF neural network to rockburst prediction based on rough set theory , 2012 .

[21]  K. Zhou,et al.  Prediction of rock burst classification using cloud model with entropy weight , 2016 .

[22]  Evert Hoek,et al.  Practical estimates of rock mass strength , 1997 .

[23]  Zhai Yinghu.  Lithology Identification Analysis Based on Normal Distribution Theory and Fuzzy Synthetic Evaluation Method , 2012 .

[24]  Li Xibing,et al.  Prediction of rockburst classification using Random Forest , 2013 .

[25]  Max Kuhn,et al.  Applied Predictive Modeling , 2013 .

[26]  T. R. Stacey,et al.  ROCKBURST MECHANISMS IN TUNNELS AND SHAFTS , 1994 .

[27]  Matthijs J. Warrens.  Cohen’s linearly weighted kappa is a weighted average , 2012, Adv. Data Anal. Classif..