基于PCA-AHPSO-SVR的煤层瓦斯含量预测研究
|
摘要
为了提高煤层瓦斯含量预测的准确性和科学性,通过主成分分析方法对影响煤层瓦斯含量的7个因素进行特征提取,消除影响因素之间的相关性,减少维度;用支持向量回归机对提取的因素进行训练,并用改进的自适应混合粒子群算法对SVR的参数进行优化,提出PCA-AHPSO-SVR模型;与PCA-PSO-SVR,PSO-SVR这2个模型在相同环境下进行30次运行比较。研究结果表明:研究提出的PCA-AHPSO-SVR模型较其他2种模型平均准确率分别提高5.51%和9.32%,稳定性更佳,可满足工程实际需求。
In order to improve the accuracy and scientificity of coal seam gas content prediction,Principal Component Analysis( PCA) is used to extract the characteristics of seven factors affecting coal seam gas content,eliminating the correlation between influencing factors and reducing the dimension,Then the support factors are trained by Support Vector Regression( SVR),and the parameters of SVR are optimized by the improved Adaptive Hybrid Particle Swarm Optimization( AHPSO),the PCA-AHPSO-SVR model was proposed and compared with the PCA-PSO-SVR and PSO-SVR models in the same environment for 30 times,the results show that the average accuracy of the model is increased by 5. 51% and 9. 32% respectively,the stability is better,and the actual needs of the project are met.
In order to improve the accuracy and scientificity of coal seam gas content prediction,Principal Component Analysis( PCA) is used to extract the characteristics of seven factors affecting coal seam gas content,eliminating the correlation between influencing factors and reducing the dimension,Then the support factors are trained by Support Vector Regression( SVR),and the parameters of SVR are optimized by the improved Adaptive Hybrid Particle Swarm Optimization( AHPSO),the PCA-AHPSO-SVR model was proposed and compared with the PCA-PSO-SVR and PSO-SVR models in the same environment for 30 times,the results show that the average accuracy of the model is increased by 5. 51% and 9. 32% respectively,the stability is better,and the actual needs of the project are met.
引文
[1]丁百川.我国煤矿主要灾害事故特点及防治对策[J].煤炭科学技术,2017,45(5):109-114.DING Baichuan.Features and prevention countermeasures of major disasters occurred in China coal mine[J].Coal Science and Technology,2017,45(5):109-114.
[2]刘艳亮.2002-2016年我国煤矿事故统计分析及预防措施[J].陕西煤炭,2018(3):64-67,40.LIU Yanliang,Statistics analysis and prevention measures of coal mine accidents in China from 2002 to 2016[J].Shaanxi Coal,2018(3):64-67,40.
[3]张子敏.瓦斯地质学[M].徐州:中国矿业大学出版社,2009.
[4]魏国营,王保军,闫江伟,等.平顶山八矿突出煤层瓦斯地质控制特征[J].煤炭学报,2015,40(3):555-561.WEI Guoying,WANG Baojun,YAN Jiangwei,et al.Gas geological control characteristics of outburst coal seam in Pingdingshan No.8mine[J].Journal of China Coal Society,2015,40(3):555-561.
[5]魏国营,门金龙,贾安立,等.基于上覆基岩特征的赵固一矿井田煤层瓦斯富集区的判识方法[J].煤炭学报,2012,37(8):1315-1319.WEI Guoying,MEN Jinlong,JIA Anli,et al.Identification method of coalbed gas enrichment area in Zhaogu No.1 mine field based on overlying bedrock characteristics[J].Journal of China Coal Society,2012,37(8):1315-1319.
[6]郝天轩,宋超.基于模糊神经网络的煤层瓦斯含量预测研究[J].中国安全科学学报,2011,21(8):36-42.HAO Tianxuan,SONG Chao.Research on prediction of coal seam gas content based on fuzzy neural network[J].China Safety Science Journal,2011,21(8):36-42.
[7]刘正,邓广哲,刘小军.多元回归分析在煤层瓦斯含量预测中的应用[J].矿业安全与环保,2013,40(5):52-55.LIU Zheng,DENG Guangzhe,LIU Xiaojun.Application of multiple regression analysis in prediction of gas content in coal seam[J].Mining Safety and Environmental Protection,2013,40(5):52-55.
[8]叶青,林柏泉.灰色理论在煤层瓦斯含量预测中的应用[J].现代矿业,2006,25(7):28-30.YE Qing,LIN Baiquan.Application of grey theory in prediction of gas content in coal seam[J].Modern Mining,2006,25(7):28-30.
[9]魏建平,郝天轩,刘明举.基于构造复杂程度定量评价的瓦斯含量预测BP模型[J].煤炭学报,2009,34(8):1090-1094.WEI Jianping,HAO Tianxuan,LIU Mingju.A BP model for predicting gas content based on quantitative evaluation of structural complexity[J].Journal of China Coal Society,2009,34(8):1090-1094.
[10]MENG Q,MA X,ZHOU Y.Forecasting of coal seam gas content by using support vector regression based on particle swarm optimization[J].Journal of Natural Gas Science&Engineering,2014,21(21):71-78.
[11]刘程程,杨力.PCA-SVR在煤层瓦斯含量预测中的应用[J].中国安全生产科学技术,2012,8(7):78-82.LIU Chengcheng,YANG Li.Application of PCA-SVR in prediction of gas content in coal seam[J].Journal of Safety Science and Technology,2012,8(7):78-82.
[12]姜谙男,梁冰,张娇.基于粒子群最小二乘支持向量机的瓦斯含量预测[J].辽宁工程技术大学学报,2009,28(3):363-366.JIANG Annan,LIANG Bing,ZHANG Jiao.Prediction of gas content based on particle swarm least squares support vector machine[J].Journal of Liaoning Technical University,2009,28(3):363-366.
[13]张尧庭,方开泰.多元统计分析引论[M].武汉:武汉大学出版社,2013.
[14]SHI Y,EBERHART R C.Parameter selection in particle swarm optimization[C]//International Conference on Evolutionary Programming Vii,Berlin:Springer-Verlag,1998:591-600.
[15]徐卫亚,徐飞,刘大文.位移时序预测的APSO-WLSSVM模型及应用研究[J].岩土工程学报,2009,31(3):313-318.XU Weiya,XU Fei,LIU Dawen.APSO-WLSSVM model for displacement time series prediction and its application[J].Chinese Journal of Geotechnical Engineering,2009,31(3):313-318.
[16]曹博,白刚,李辉.基于PCA-GA-BP神经网络的瓦斯含量预测分析[J].中国安全生产科学技术,2015,11(5):84-90.CAO Bo,BAI Gang,LI Hui.Prediction of gas content based on PCA-GA-BP neural network[J].Journal of Safety Science and Technology,2015,11(5):84-90.
[17]彭立世,袁崇孚.瓦斯地质与瓦斯突出预测[M].北京:中国科学技术出版社,2009.
[2]刘艳亮.2002-2016年我国煤矿事故统计分析及预防措施[J].陕西煤炭,2018(3):64-67,40.LIU Yanliang,Statistics analysis and prevention measures of coal mine accidents in China from 2002 to 2016[J].Shaanxi Coal,2018(3):64-67,40.
[3]张子敏.瓦斯地质学[M].徐州:中国矿业大学出版社,2009.
[4]魏国营,王保军,闫江伟,等.平顶山八矿突出煤层瓦斯地质控制特征[J].煤炭学报,2015,40(3):555-561.WEI Guoying,WANG Baojun,YAN Jiangwei,et al.Gas geological control characteristics of outburst coal seam in Pingdingshan No.8mine[J].Journal of China Coal Society,2015,40(3):555-561.
[5]魏国营,门金龙,贾安立,等.基于上覆基岩特征的赵固一矿井田煤层瓦斯富集区的判识方法[J].煤炭学报,2012,37(8):1315-1319.WEI Guoying,MEN Jinlong,JIA Anli,et al.Identification method of coalbed gas enrichment area in Zhaogu No.1 mine field based on overlying bedrock characteristics[J].Journal of China Coal Society,2012,37(8):1315-1319.
[6]郝天轩,宋超.基于模糊神经网络的煤层瓦斯含量预测研究[J].中国安全科学学报,2011,21(8):36-42.HAO Tianxuan,SONG Chao.Research on prediction of coal seam gas content based on fuzzy neural network[J].China Safety Science Journal,2011,21(8):36-42.
[7]刘正,邓广哲,刘小军.多元回归分析在煤层瓦斯含量预测中的应用[J].矿业安全与环保,2013,40(5):52-55.LIU Zheng,DENG Guangzhe,LIU Xiaojun.Application of multiple regression analysis in prediction of gas content in coal seam[J].Mining Safety and Environmental Protection,2013,40(5):52-55.
[8]叶青,林柏泉.灰色理论在煤层瓦斯含量预测中的应用[J].现代矿业,2006,25(7):28-30.YE Qing,LIN Baiquan.Application of grey theory in prediction of gas content in coal seam[J].Modern Mining,2006,25(7):28-30.
[9]魏建平,郝天轩,刘明举.基于构造复杂程度定量评价的瓦斯含量预测BP模型[J].煤炭学报,2009,34(8):1090-1094.WEI Jianping,HAO Tianxuan,LIU Mingju.A BP model for predicting gas content based on quantitative evaluation of structural complexity[J].Journal of China Coal Society,2009,34(8):1090-1094.
[10]MENG Q,MA X,ZHOU Y.Forecasting of coal seam gas content by using support vector regression based on particle swarm optimization[J].Journal of Natural Gas Science&Engineering,2014,21(21):71-78.
[11]刘程程,杨力.PCA-SVR在煤层瓦斯含量预测中的应用[J].中国安全生产科学技术,2012,8(7):78-82.LIU Chengcheng,YANG Li.Application of PCA-SVR in prediction of gas content in coal seam[J].Journal of Safety Science and Technology,2012,8(7):78-82.
[12]姜谙男,梁冰,张娇.基于粒子群最小二乘支持向量机的瓦斯含量预测[J].辽宁工程技术大学学报,2009,28(3):363-366.JIANG Annan,LIANG Bing,ZHANG Jiao.Prediction of gas content based on particle swarm least squares support vector machine[J].Journal of Liaoning Technical University,2009,28(3):363-366.
[13]张尧庭,方开泰.多元统计分析引论[M].武汉:武汉大学出版社,2013.
[14]SHI Y,EBERHART R C.Parameter selection in particle swarm optimization[C]//International Conference on Evolutionary Programming Vii,Berlin:Springer-Verlag,1998:591-600.
[15]徐卫亚,徐飞,刘大文.位移时序预测的APSO-WLSSVM模型及应用研究[J].岩土工程学报,2009,31(3):313-318.XU Weiya,XU Fei,LIU Dawen.APSO-WLSSVM model for displacement time series prediction and its application[J].Chinese Journal of Geotechnical Engineering,2009,31(3):313-318.
[16]曹博,白刚,李辉.基于PCA-GA-BP神经网络的瓦斯含量预测分析[J].中国安全生产科学技术,2015,11(5):84-90.CAO Bo,BAI Gang,LI Hui.Prediction of gas content based on PCA-GA-BP neural network[J].Journal of Safety Science and Technology,2015,11(5):84-90.
[17]彭立世,袁崇孚.瓦斯地质与瓦斯突出预测[M].北京:中国科学技术出版社,2009.