基于核主成分分析与SVM的岩爆烈度组合预测模型

doi:10.11872/j.issn.1005-2518.2020.04.019

Abstract

Abstract:

Rockburst is a relatively dangerous engineering geological disaster in underground hard rock engineering constructed in high geostress area.Due to the re-distribution of the stress in surrounding rocks during the excavation of underground engineering，the elastic strain energy is released suddenly and abruptly，causing rock fragments to eject from the rock.And then，the casualties and equipment damage are often happened，which make the rockburst become one of the worldwide difficulties in underground engineering.Therefore，the prediction of possibility of rockburst and its intensity is a problem that must be solved in underground engineering construction.For predicting rock-burst intensity effectively，a combined prediction model based on kernel principal component analysis （KPCA） of multiple types and the support vector machine （SVM） optimized by genetic algorithm or particle swarm optimization algorithm （GA/PSO） was established.According to the characteristics and causes of rockburst，rocks’ maximum tangential stress $σ θ$ ，rocks’ uniaxial compressive strength $σ t$ ，rocks’ uniaxial tensile strength $σ c$ ，stress concentration coefficient $S C F$ ，rock brittleness coefficient $B 1$ and $B 2$ ，and elastic energy index $W e t$ were chosen to form the rockburst prediction indexes system.Based on 246 groups of typical rockburst cases at home and abroad，the data were preprocessed through the principal component analysis and the principal component analysis based on linear kernel function，radial basis function （RBF） kernel function and multi-layer perceptron （MLP） kernel function.On the basis of ensuring the amount of information in the original data，2 to 4 linearly independent principal components are obtained，which reduces the correlation between the indicators and the input parameters of the SVM model，and simplifies the training process.Then input the dimensionality-reduced data into GA/PSO optimized SVM model for training and prediction.To improve classification accuracy and generalization ability of the SVM，GA/PSO were adopted to automatically determine the parameters for support vector machine，and the optimal values of parameters $C$ and $g$ were determined by the method of 10 fold cross validation，which avoided the blindness of manually providing parameters.In this study，220 rockburst samples were randomly selected as the training set，and the remaining 26 samples were selected as the test set.After testing，the optimal parameters，the training set and test set accuracy of the 8 combined models were obtained.The prediction accuracy of the model based on the combination of the principal component analysis method of RBF kernel function and PSO-SVM reached 92.3%，which was the optimal combination model.It demonstrated that the combined prediction model can accurately deal with the complex non-linear relationship between various factors affecting the rockburst intensity，and the model has strong engineering practicability in the prediction of rockburst intensity.

Key words: kernel principal component analysis, prediction of rockburst intensity, genetic algorithm, particle swarm optimization algorithm, support vector machine, combination prediction model

CLC Number:

TU45

Rui XU, Kuikui HOU, Xi WANG, Xingquan LIU, Xibing LI. Combined Prediction Model of Rockburst Intensity Based on Kernel Principal Component Analysis and SVM[J].Gold Science and Technology, 2020, 28(4): 575-584.

Figures/Tables 12

Fig.1

Table 1

Partial raw data of rockburst samples"

序号	$σ θ$ /MPa	$σ c$ /MPa	$σ t$ /MPa	SCF	B₁	B₂	$W e t$	岩爆烈度
1	90.00	170.00	11.30	0.53	15.04	0.88	9.00	M
2	90.00	220.00	7.40	0.41	29.73	0.93	7.30	L
3	62.60	165.00	9.40	0.38	17.53	0.89	9.00	L
4	55.40	176.00	7.30	0.32	24.11	0.92	9.30	M
5	30.00	88.70	3.70	0.34	23.97	0.92	6.60	M
6	48.75	180.00	8.30	0.27	21.69	0.91	5.00	M
$?$	$?$	$?$	$?$	$?$	$?$	$?$	$?$	$?$
243	126.72	189.70	8.95	0.67	21.20	0.91	5.43	L
244	57.97	125.37	7.74	0.67	21.20	0.46	2.86	L
245	57.97	96.16	3.77	0.46	16.20	0.20	2.53	L
246	57.97	70.68	4.19	0.60	25.51	0.19	2.87	L

Table 1

Fig.2

Table 2

Correlation coefficient of rockburst evaluation indexes"

指标	$σ θ$ /MPa	$σ c$ /MPa	$σ t$ /MPa	SCF	B₁	B₂	$W e t$
$σ θ$ /MPa	1.000	0.034	0.334	0.918	-0.248	-0.219	0.468
$σ c$ /MPa	0.034	1.000	0.422	-0.265	0.076	0.234	0.193
$σ t$ /MPa	0.334	0.422	1.000	0.164	-0.625	-0.471	0.329
SCF	0.918	-0.265	0.164	1.000	-0.258	-0.252	0.329
B₁	-0.248	0.076	-0.625	-0.258	1.000	0.526	-0.108
B₂	-0.219	0.234	-0.471	-0.252	0.526	1.000	-0.082
$W e t$	0.468	0.193	0.329	0.329	-0.108	-0.082	1.000

Table 2

Table 3

Results for the standardized processing of raw data"

序号	$σ θ$ /MPa	$σ c$ /MPa	$σ t$ /MPa	SCF	B₁	B₂	$W e t$
1	0.5917	1.3704	0.9878	-0.0869	-0.3864	0.1446	0.9237
2	0.5917	2.5425	0.0553	-0.2655	0.6451	0.6570	0.5147
3	0.0850	1.2532	0.5335	-0.3101	-0.2116	0.2470	0.9237
4	-0.0482	1.5111	0.0314	-0.3994	0.2505	0.5545	0.9959
5	-0.5179	-0.5353	-0.8294	-0.3697	0.2407	0.5545	0.3463
6	-0.1711	1.6049	0.2705	-0.4738	0.0805	0.4520	-0.0387
$?$	$?$	$?$	$?$	$?$	$?$	$?$	$?$
243	1.2708	1.8322	0.4259	0.1215	0.0461	0.4520	0.0648
244	-0.0006	0.3243	0.1366	0.1215	0.0461	-4.1597	-0.5536
245	-0.0006	-0.3604	-0.8126	-0.1911	-0.3050	-6.8242	-0.6330
246	-0.0006	-0.9577	-0.7122	0.0173	0.3488	-6.9267	-0.5512

Table 3

Fig.3

Fig.4

Table 4

Selection of kernel parameters and principal components"

核函数类型	核函数表达式	核参数	第一主成分贡献率/ $%$	主成分个数
主成分分析	-	-	40.20	4
线性核函数	$K x i, x j = x i ? x j$	-	59.19	2
高斯核函数	$K x i, x j = e x p ? - x i - x j 2 2 σ 2$	$σ = 890$	58.52	3
多层感知器核函数	$K x i, x j = t a n h ? v x i ? x j + c$	$v = 1 × 10 - 5, c = - 5$	61.59	3

Table 4

Fig.5

Table 5

Results for the SVM parameters，training set accuracy and test set accuracy"

模型		SVM参数		训练集准确率/ $%$	测试集准确率/ $%$
模型		$C$	$g$	训练集准确率/ $%$	测试集准确率/ $%$
PCA	GA-SVM	81.4710	56.6212	96.8	84.6
PCA	PSO-SVM	97.6345	118.4431	98.2	84.6
KPCA1	GA-SVM	50.5654	890.2415	94.6	73.1
KPCA1	PSO-SVM	51.3095	586.1588	89.6	76.9
KPCA2	GA-SVM	3.1284	256.1905	91.4	88.5
KPCA2	PSO-SVM	31.2537	145.4178	94.6	92.3
KPCA3	GA-SVM	73.1933	282.2821	93.2	80.8
KPCA3	PSO-SVM	82.0614	491.9322	97.7	88.5

Table 5

Fig.6

Fig.7

References 19

1	李夕兵.岩石动力学基础与应用[M].北京：科学出版社，2014.
	Li Xibing.Rock Dynamics Fundamentals and Applications[M].Beijing：Science Press，2014.
2	王元汉，李卧东，李启光，等.岩爆预测的模糊数学综合评判方法[J].岩石力学与工程学报，1998，17（5）：493-501.
	Wang Yuanhan，Li Wodong，Li Qiguang，et al.Method of fuzzy comprehensive evaluations for rockburst prediction[J].Chinese Journal of Rock Mechanics and Engineering，1998，17（5）：493-501.
3	李夕兵，宫凤强，王少锋，等.深部硬岩矿山岩爆的动静组合加载力学机制与动力判据[J].岩石力学与工程学报，2019，38（4）：708-723.
	Li Xibing，Gong Fengqiang，Wang Shaofeng，et al.Coupled static-dynamic loading mechanical mechanism and dynamic criterion of rockburst in deep hard rock mines[J].Chinese Journal of Rock Mechanics and Engineering，2019，38（4）：708-723.
4	宫凤强，李夕兵.岩爆发生和烈度分级预测的距离判别方法及应用[J].岩石力学与工程学报，2007，26（5）：1012-1018.
	Gong Fengqiang，Li Xibing.A distance discriminant analysis method for prediction of possibility and classification of rockburst and its application[J].Chinese Journal of Rock Mechanics and Engineering，2007，26（5）：1012-1018.
5	赵洪波.岩爆分类的支持向量机方法[J].岩土力学，2005，26（4）：642-644.
	Zhao Hongbo.Classification of rockburst using support vector machine[J].Rock and Soil Mechanics，2005，26（4）：642-644.
6	白明洲，王连俊，许兆义.岩爆危险性预测的神经网络模型及应用研究[J].中国安全科学学报，2002，12（4）：65-69.
	Bai Mingzhou，Wang Lianjun，Xu Zhaoyi.Study on a neutral network model and its application in predicting the risk of rock blast[J].China Safety Science Journal，2002，12（4）：65-69.
7	高玮.基于蚁群聚类算法的岩爆预测研究[J].岩土工程学报，2010，32（6）：874-880.
	Gao Wei.Prediction of rock burst based on ant colony clustering algorithm[J].Chinese Journal of Geotechnical Engineering，2010，32（6）：874-880.
8	Zhou J，Li X B，Shi X Z.Long-term prediction model of rockburst in underground openings using heuristic algorithms and support vector machines[J].Safety Science，2012，50（4）：629-644.
9	邱道宏，李术才，张乐文，等.基于模型可靠性检查的QGA-SVM岩爆倾向性分类研究[J].应用基础与工程科学学报，2015，23（5）：981-991.
	Qiu Daohong，Li Shucai，Zhang Lewen，et al.Research on QGA-SVM rock burst orientation classification based on model reliability examination[J].Journal of Basic Science and Engineering，2015，23（5）：981-991.
10	贾义鹏，吕庆，尚岳全.基于粒子群算法和广义回归神经网络的岩爆预测[J].岩石力学与工程学报，2013，32（2）：344-348.
	Jia Yipeng，Qing Lü，Shang Yuequan.Rockburst prediction using particle swarm optimization algorithm and general regression neural network[J].China Journal of Rock Mechanics and Engineering，2013，32（ 2）：343-348.
11	Liu S，Dou L M，Si G Y，et al.A principal component analysis/fuzzy comprehensive evaluation model for coal burst liability assessment[J].International Journal of Rock Mechanics and Mining Sciences，2016，81（5）：62-69.
12	吴顺川，张晨曦，成子桥.基于PCA-PNN原理的岩爆烈度分级预测方法[J].煤炭学报，2019，44（9）：2767-2776.
	Wu Shunchuan，Zhang Chenxi，Cheng Ziqiao. Prediction of intensity classification of rockburst based on PCA-PNN principle[J].Journal of China Coal Society，2019，44（9）：2767-2776.
13	Zhou J，Li X B，Hani S M.Classification of rockburst in underground projects comparison of ten supervised learning methods[J].Journal of Computing in Civil Engineering，2016，30（5）：1-19.
14	赵国彦，刘雷磊，王剑波，等.岩爆等级预测的PCA-OPF模型[J].矿冶工程，2019，39（4）：1-5.
	Zhao Guoyan，Liu Leilei，Wang Jianbo，et al.PCA-OPF model for rock burst prediction[J].Mining and Metallurgical Engineering，2019，39（4）：1-5.
15	辛焕平.MATLAB R2017a模式识别与智能计算[M].北京：电子工业出版社，2018.
	Xin Huanping.MATLAB R2017a Pattern Recognition and Intelligent Computing[M].Beijing：Publishing House of Electronics Industry，2018.
16	Sun H T，Lü G D，Mo J Q，et al.Application of KPCA combined with SVM in Raman spectral discrimination[J].Optik，2019，184（5）：214-219.
17	冯夏庭，赵洪波.岩爆预测的支持向量机[J].东北大学学报（自然科学版），2002，23（1）：57-59.
	Feng Xiating，Zhao Hongbo.Prediction of rockburst using support vector machine［J］.Journal of Northeastern University（Natural Science），2002，23（ 1）：57-59.
18	陈晋音，熊晖，郑海斌.基于粒子群算法的支持向量机的参数优化[J].计算机科学，2018，45（6）：197-203.
	Chen Jinyin，Xiong Hui，Zheng Haibin.Parameters optimization for SVM based on particle swarm algorithm[J].Computer Science，2018，45（6）：197-203.
19	Pu Y Y，Apel D B，Xu H W.Rockburst prediction in kimberlite with unsupervised learning method and support vector classifier[J].Tunnelling and Underground Space Technology，2019，90：12-18.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

Comments

Recommended 10

[1]	YI Cun-Chang, CANG En-Guang. [J]. J4, 2010, 18(4): 58 -61 .
[2]	LU Ren-Jiang, LIU Feng-Jin, LOU Wei-Hua-. Technology Innovation and Application on High Water Consolidation of Fill Mining Method[J]. J4, 2011, 19(1): 51 -54 .
[3]	ZHANG YinFei. Discussion on M ineralization Regularity of the Eighth Ore Body of Shuangwang Gold Deposit, Shaanx i Prov ince[J]. J4, 2008, 16(1): 35 -38 .
[4]	YANG Meng-Rong, MAO Chang-Xian. Uncertainty Evaluation of Arsenic and Antimony in Chemical Prospecting Sa-mple by Atomic Fluorescence Spectrometry[J]. J4, 2010, 18(3): 68 -71 .
[5]	GENG A-Jiao, DUAN Jian-Hua. Analysis on the Ore-controlling Factors and Prospecting of Manzhanggang Gold Mine[J]. J4, 2010, 18(6): 34 -37 .
[6]	KANG Meng，MA Meng-Hao. Application Research on Electrogeochemical Method for Regional Exploration[J]. J4, 2008, 16(1): 1 -6 .
[7]	XIE Guiming. Thought of Concern ingM in ing Right Va lue andM in ing Right Sun of Money[J]. J4, 2008, 16(1): 58 -61 .
[8]	ZANG Enguang，YI Cunchang，ZHANG Chunxiao. The Geologic Feature and the Clue for Prospecting of Huana Conglomerate Gold Deposit in Heilongjiang[J]. J4, 2008, 16(2): 29 -32 .
[9]	ZHUANG Yukai，XU Zhongmin，FEN G Jinmin，LU Liangshan，LUAN Zuochun. Acid and Alkali Absorb the Acid Gas in the W et Smelt System[J]. J4, 2008, 16(2): 57 -59 .
[10]	HUANG Jian-Jun, LI Tian-En, FAN Gong-Ke. Discussion on Geological Setting of Metallogenic and Prospecting Potentiality of Daxinganling Region[J]. J4, 2010, 18(6): 13 -17 .

Combined Prediction Model of Rockburst Intensity Based on Kernel Principal Component Analysis and SVM

RichHTML

PDF (PC)

Abstract

Cite this article

share this article

Figures/Tables 12

References 19

Related Articles 6

Metrics

Comments

Recommended 10

[1]	Renhao LI,Helong GU,Xibing LI,Kuikui HOU,Deming ZHU,Xi WANG. A PSO-RBF Neural Network Model for Rockburst Tendency Prediction [J]. Gold Science and Technology, 2020, 28(1): 134-141.
[2]	Chaoqun HAN,Jianhong CHEN,Zhiyong ZHOU,Shan YANG. Research on Prediction of Rock Mass Blastability Classification Based on PCA-SVM Model [J]. Gold Science and Technology, 2019, 27(6): 879-887.
[3]	Wenfeng XIAO,Jianhong CHEN,Yi CHEN,Ximei WANG. Optimization of Multi-objective Filling Slurry Ratio Based on Neural Network and Genetic Algorithm [J]. Gold Science and Technology, 2019, 27(4): 581-588.
[4]	ZHANG Qinli，WANG Jing，WANG Xinmin. Failure Risk Assessment Model of Filling Pipeline Based on KPCA and PSO-SVM [J]. Gold Science and Technology, 2017, 25(3): 70-76.
[5]	LIU Zhixiang，LIU Yiran，LAN Ming. Research on PCA-GA-ELM Model of Mine Inflow Prediction [J]. Gold Science and Technology, 2017, 25(1): 61-67.
[6]	YE Pei. Application of a New Quantum Genetic Algorithm on Denoising Processing of GPS Satellite Signal at the Yangshan Gold Deposit [J]. Gold Science and Technology, 2015, 23(2): 83-87.