为快速、准确地预测冲击地压危险性,借鉴随机森林理论,选取影响冲击地压的10 项主要因素:煤层、倾角、埋深、构造情况、倾角变化、煤厚变化、瓦斯浓度、顶板管理、卸压、响煤炮声作为判别因子,建立冲击地压危险性识别的随机森林模型.利用重庆砚石台矿24 组实测数据作为学习样本建立随机森林分类器,在对样本分类的同时,计算预测变量的重要性值GI,发现构造情况为最重要的评价指标,其后是响煤炮声和倾角.利用其他12 组现场数据作为预测样本对该模型进行测试,预测结果与实际情况吻合较好.
Arandom forest (RF) modelfor rock burst identification was established on the basis of the RF theory to forecast rock burst risk rapidly and accurately. Ten indices, ie, coal seam, dip angle, buried depth, structure situation, change of dip angle, change of coal thickness, gas concentration, roof management, pressure relief and shooting were used as the criterion indices for rock burst prediction in the proposed model on the basis of analysis of rock burst impact. Twenty-four typical rock burst instances of a coal mine were used to createa RF classifier. RF is a combination of tree predictors, and variable importance is measured by Gini importance (GI) when the forest grows. The GI shows that structure situation was the most important indicator, followed by shooting and dip angle. Another 12 groups of rock burst instances were tested as forecast samples, and the predicted results were in accordance with actual situation.
[1] 窦林名, 何学秋. 冲击地压防治理论与技术[M]. 徐州: 中国矿业大学 出版社, 2001. Dou Linming, He Xueqiu. Theory and technology of rock burst prevention[M]. Xuzhou: China University of Mining and Technology Press, 2001.
[2] 姚精明, 何富连, 徐军, 等. 冲击地压的能量机理及其应用[J]. 中南大 学学报: 自然科学版, 2009, 40(3): 808-813. Yao Jingming, He Fulian, Xu Jun, et al. Energy mechanism of rock burst and its application[J]. Journal of Central South University: Natural Science Edition, 2009, 40(3): 808-813.
[3] 段宏飞, 姜振泉, 朱术云, 等. 大埋深高地压弱结构面顶板岩层诱发冲 击地压离心模型试验[J]. 中南大学学报: 自然科学版, 2011, 42(9): 2774-2782. Duan Hongfei, Jiang Zhenquan, Zhu Shuyun, et al. Centrifuge model tests on rock brusting induced by great depth highly stressed roof strata of weak structural plane[J]. Journal of Central South University: Natural Science Edition, 2011, 42(9): 2774-2782.
[4] 国家煤矿安全监察局."十一五"期间(2006—2010年)全国煤矿事故 分析报告[R]. 北京: 国家煤矿安全监察局, 2011. China Administration of Coal Mine Safety. Report of the national coal mine accidents in"Eleven Five" period (2006—2010) [R]. Beijing: China Administration of Coal Mine Safety, 2011.
[5] 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.
[6] 李文, 纪洪广, 武玉梁. 深井冲击地压发生机理分析及预测方法研究[J]. 中国矿业, 2007, 16(7): 105-107． Li Wen, Ji Hongguan, Wu Yulian. Mechanism analyses and forecast on deep mining rock burst[J]. China Mining Magazine, 2007, 16(7): 105- 107.
[7] 陈国祥, 窦林名, 曹安业, 等. 电磁辐射法评定冲击矿压危险等级及应 用[J]. 煤炭学报, 2008, 33(8): 866-870. Chen Guoxiang, Dou Linming, Cao Anye, et al. Assessment of rock burst danger and application on electromagnetic emission method[J]. Journal of China Coal Society, 2008, 33(8): 866-870.
[8] 李洪, 戴仁竹, 蒋金泉. 基于最大Lyapunov 指数的冲击地压预测模型[J]. 采矿与安全工程学报, 2006, 23(2): 215-219. Li Hong, Dai Renzhu, Jiang Jinquan. Predicting model of rock burst based on Lyapunov index[J]. Journal of Mining & Safety Engineering, 2006, 23(2): 215-219.
[9] 赵洪波, 茹忠亮. 冲击地压预测的PSO-SVM模型[J]. 岩石力学与工程 学报, 2007, 26(增1): 3479-3483. Zhao Hongbo, Ru Zhongliang. PSO-SVM model for prediction of rock burst[J]. Chinese Journal of Rock Mechanics and Engineering, 2007, 26 (Suppl 1): 3479-3483.
[10] 尹光志, 谭钦文, 魏作安. 基于混沌优化神经网络的冲击地压预测模 型[J]. 煤炭学报, 2008, 33(8): 871-875. Yin Guangzhi, Tan Qinwen, Wei Zuoan. Combined optimization model of rock-burst prediction based on chaos optimization and BP neural networks[J]. Journal of China Coal Society, 2008, 33(8): 871-875.
[11] 闫河. 基于遗传算法和人工神经网络相结合的冲击地压预测的研究[D]. 重庆: 重庆大学, 2002. Yan He. The research of rockburst prediction basing on genetic algorithm and bp artificial neural network[D]. Chongqing: Chongqing University, 2002.
[12] 周健, 史秀志. 冲击地压危险性等级预测的Fisher判别分析方法[J]. 煤炭学报, 2010, 35(增1): 22-27. Zhou Jian, Shi Xiuzhi. Fisher discriminant analysis method for prediction of classification of rock burst risk[J]. Journal of China Coal Society, 2010, 35(Suppl 1): 22-27.
[13] 金佩剑, 王恩元, 刘晓斐. 冲击地压危险性综合评价的突变级数法研 究[J]. 采矿与安全工程学报, 2013, 30(2): 256-261. Jin Peijian, Wang Enyuan, Liu Xiaofei. Catastrophe progression method on comprehensive evaluation of rock burst[J]. Journal of Mining & Safety Engineering, 2013, 30(2): 256-261.
[14] Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.
[15] 方匡南, 吴见彬, 朱建平, 等. 随机森林方法研究综述[J]. 统计与信 息论坛, 2011, 26(3): 32-38. Fang Kuangnan, Wu Jianbin, Zhu Jianping, et al. A review of technologies on random forests[J]. Statistics & Information Forum, 2011, 26(3): 32-38.
[16] Liaw A, Wiener M. Classification and regression by random forest[J]. R News, 2002, 2(3): 118-122.
[17] 秦锋, 杨波, 程泽凯. 分类器性能评价标准研究[J]. 计算机技术与发 展, 2006, 16(10): 85-88. Qin Feng, Yang Bo, Chen Zekai. Research on measure criteria in evaluating classification performance[J]. Computer Technology and Development, 2006, 16(10): 85-88.
