基于RS-PCA-GA-SVM的砂土液化预测方法研究
|
摘要
砂土液化是一种危害性比较大的自然灾害,对砂土液化进行判定预测在地质灾害防治领域中有重要的研究意义。通过粗糙集理论(Rough Set,RS)对影响砂土液化的6个初始评价指标(包括震级、土深、震中距、地下水位、标贯击数和地震持续时间)进行属性约简,去掉冗余或干扰信息,得到基于4个核心预测指标的数据集。通过主成分分析法(Principal Component Analysis,PCA)从核心评价指标中提取出主成分,采用支持向量机(Support Vector Machine,SVM)对数据集进行训练,用遗传算法(Genetic Algorithm,GA)优化参数,建立砂土液化的RS-PCA-GA-SVM预测模型。并结合砂土液化实际数据将预测结果与基于Levenberg-Marquardt算法改进的BP神经网络模型(LM-BP)的预测结果做比较。实例计算表明:基于RS-PCA-GA-SVM模型得到的砂土液化预测结果精度较LM-BP神经网络有很大的提高,判别结果与实际情况比较吻合,可在实际工程中应用。
Sand liquefaction is a harmful natural disaster, and it is of great importance to evaluate and predict sand liquefaction in the field of geological disaster prevention and control. In this paper, the rough set theory(RS) was used to perform attribute reduction on six initial evaluation indices, including magnitude, depth of soil, epicentral distance, groundwater level, standard penetration test blow count, and earthquake duration, all of which affect sand liquefaction. After removing redundant or interference information, we obtained a data set based on four core predictors. The principal component analysis(PCA) method was then used to extract the principal component from the four-core evaluation indices. The support vector machine(SVM) was used to train the data set, and the genetic algorithm(GA) was used to optimize the parameters. Finally, the RS-PCA-GA-SVM prediction model for sand liquefaction was established. Combined with the actual data of sand liquefaction, the predicted result of the proposed model was compared with that of the back propagation(BP) neural network model based on the improved Levenberg-Marquardt algorithm(LM-BP). The calculated results showed that the accuracy of sand liquefaction prediction results based on a RS-PCA-GA-SVM model are much better than those of the LM-BP neural network. The discriminant results were in good agreement with the actual situation and can be applied in practical engineering.
Sand liquefaction is a harmful natural disaster, and it is of great importance to evaluate and predict sand liquefaction in the field of geological disaster prevention and control. In this paper, the rough set theory(RS) was used to perform attribute reduction on six initial evaluation indices, including magnitude, depth of soil, epicentral distance, groundwater level, standard penetration test blow count, and earthquake duration, all of which affect sand liquefaction. After removing redundant or interference information, we obtained a data set based on four core predictors. The principal component analysis(PCA) method was then used to extract the principal component from the four-core evaluation indices. The support vector machine(SVM) was used to train the data set, and the genetic algorithm(GA) was used to optimize the parameters. Finally, the RS-PCA-GA-SVM prediction model for sand liquefaction was established. Combined with the actual data of sand liquefaction, the predicted result of the proposed model was compared with that of the back propagation(BP) neural network model based on the improved Levenberg-Marquardt algorithm(LM-BP). The calculated results showed that the accuracy of sand liquefaction prediction results based on a RS-PCA-GA-SVM model are much better than those of the LM-BP neural network. The discriminant results were in good agreement with the actual situation and can be applied in practical engineering.
引文
[1] 高大钊,袁聚云.土质学与土力学[M].北京:人民交通出版社,2001:216-222.GAO Dazhao,YUAN Juyun.Soil Science and Soil Mechanics[M].Beijing:China Communications Press,2001:216-222.
[2] 李波,苏经宇,马东辉,等.地震砂土液化判别的灰色关联-逐步分析耦合模型[J].中南大学学报(自然科学版),2016,47(1):232-238.LI Bo,SU Jingyu,MA Donghui,et al.Coupling Model Based on Grey Relational Analysis and Stepwise Discriminant Analysis for Seismic Liquefaction Discrimination of Sandy Soil[J].Journal of Central South University (Science and Technology),2016,47(1):232-238.
[3] 刘年平,王宏图,袁志刚,等.砂土液化预测的Fisher判别模型及应用[J].岩土力学,2012,33(2):554-546.LIU Nianping,WANG Hongtu,YUAN Zhigang,et al.Fisher Discriminant Analysis Model and Its Application to Sand Liquefaction Prediction[J].Hydrogeology & Engineering Geology,2012,33(2):554-546.
[4] 祝百茹,刘海卿.基于模糊综合评判的砂土液化判别[J].广西大学学报(自然科学版),2013,38(1):144-150.ZHU Bairu,LIU Haiqing.Fuzzy Comprehensive Evaluation on Liquefaction of SandSoil[J].Journal of Guangxi University (Nat Sci Ed),2013,38(1):144-150.
[5] 高宗军,付青,郑秋霞,等.BP和Elman神经网络在砂土液化预测中的研究[J].中国安全生产科学技术,2013,9(6):58-62.GAO Zongjun,FU Qing,ZHENG Qiuxia,et al.Study on Forecasting of Sand Liquefaction by Using BP Neural and Elamn Neural Networks[J].Journal of Safety Science and Technology,2013,9(6):58-62.
[6] 李振林.砂土地震液化判别方法的综合应用[J].西安科技大学学报,2010,30(4):451-456.LI Zhenlin.Discrimination Methods for Sandy Soil Seismic Liquefaction[J].Journal of Xi’an University of Science and Technology,2010,30(4):451-456.
[7] 夏建中,罗战友,龚晓南,等.基于支持向量机的砂土液化预测模型[J].岩石力学与工程学报,2005,24(22):4139-4144.XIA Jianzhong,LUO Zhanyou,GONG Xiaonan,et al.Support Vector Machine Model for Predicting Sand Liquefaction[J].Chinese Journal of Rock Mechanics and Engineering,2005,24(22):4139-4144.
[8] 师旭超,郭志涛,韩阳.基于支持向量机的砂土液化预测分析[J].地震工程学报,2009,31(4):363-366.SHI Xuchao,GUO Zhitao,HAN Yang.Analysis on Sand Seismic Liquefaction Prediction Based on the Support Vector Machine[J].China Earthquake Engineering Journal,2009,31(4):363-366.
[9] 张向东,冯胜洋,王长江.基于网格搜索的支持向量机砂土液化预测模型[J].应用力学学报,2011,28(1):24-28.ZHANG Xiangdong,FENG Shengyang,WANG Changjiang.Support Vector Machine Model for Predicting Sand Liquefaction Based on Grid-Search Method[J].Chinese Journal of Applied Mechanics,2011,28(1):24-28.
[10] WANG F,SU J,WANG Z.Forecasting of Building Settlements Due to Earthquake Liquefaction Based on LS-SVM with Mixed Kernel[J].Electronic Journal of Geotechnical Engineering,2015,20(1):11-19.
[11] FAN Z J,LENG Y Q,XU Y L,et al.A Discrimination Method of Saturated Sand Liquefaction Possibility Based on Support Vector Machine[J].Applied Mechanics & Materials,2014,509:38-43.
[12] PAWLAK Z,Rough Set[J].International Journal of Computer and Information Sciences,1982,11:341-356.
[13] 高爽,冬雷,高阳,等.基于粗糙集理论的中长期风速预测[J].中国电机工程学报,2012,32(1):32-37.GAO Shuang,DONG Lei,GAO Yang,et al.Mid-long Term Wind Speed Prediction Based on Rough Set Theory[J].Proceedings of the CSEE,2012,32(1):32-37.
[14] 张文修,吴伟志.粗糙集理论介绍和研究综述[J].模糊系统与数学,2000,14(4):1-12.ZHANG Wenxiu,WU Weizhi.An Introduction and a Survey for the Studies of Rough Set Theory[J].Fuzzy Systems and Mathematics,2000,14(4):1-12.
[15] 张国英,王娜娜,张润生,等.基于主成分分析的BP神经网络在岩性识别中的应用[J].北京石油化工学院学报,2008,16(3):43-46.ZHANG Guoying,WANG Nana,ZHANG Runsheng,et al.Application of Principal Component Analysis and BP Neural Network in Identifying Lithology[J].Journal of Beijing Institute of Petro-chemical Technology,2008,16(3):43-46.
[16] VAPNIK V N.The Nature of Statistical Learning Theory[M].New York:SpringVerlag,1995.
[17] 杨健,陈庆寿.砂土液化影响因素及其判别方法[J].西部探矿工程,2004,93(2):1-2.YANG Jian,CHEN Qingshou.Affecting Factors and Distinguishing Methods of Sandy Soil Liquefaction[J].West-China Exploration Engineering,2004,93(2):1-2.
[18] 蔡煜东,宫家文.砂土液化预测的人工神经网络模型[J].岩土工程学报,1993,15(6):53-58.CAI Yudong,GONG Jiawen.Artificial Neural Network Model for Prediction of Sand Liquefaction[J].Chinese Journal of Geotechnical Engineering,1993,15(6):53-58.
[19] 王举范,陈卓.基于信息熵的粗糙集连续属性多变量离散化算法[J].青岛科技大学学报(自然科学版),2013,34(4):423-426.WANG Jufan,CHEN Zhuo.Multiple Variable Discretization Algorithm of Continuous Attributes in Rough Set Theory Based on Information Entropy[J].Journal of Qingdao University of Science and Technology (Natural Science Edition),2013,34(4):423-426.
[20] WONG S K M,ZIARKO W.On Optiongal Decision Rule in Decision Table[J].Bulletin of Polish Academy of Science,1985,33:693-696.
[21] 郭超,宋卫华,魏威.基于网格搜索-支持向量机的采场顶板稳定性预测[J].中国安全科学学报,2014,24(8):31-36.GUO Chao,SONG Weihua,WEI Wei.Stope Roof Stability Prediction Based on Both SVM and Grid-search Method[J].China Safety Science Journal,2014,24(8):31-36.
[22] 谢玮,王彦春,刘建军,等.基于粒子群优化最小二乘支持向量机的非线性AVO反演[J].石油地球物理勘探,2016,51(6):1187-1194.XIE Wei,WANG Yanchun,LIU Jianjun,et al.Non-linear AVO Inversion Based on PSO-LSSVM[J].Oil Geophysical Prospecting,2016,51(6):1187-1194.
[23] 季斌,周涛发,袁峰.遗传算法优化支持向量机矿产预测方法[J].测绘科学,2015,40(10):106-109.JI Bin,ZHOU Taofa,YUAN Feng.The Detection Method of Maglev Gyroscope Abnormal Data Based on the Characteristics of Two Positioning[J].Science of Surveying and Mapping,2015,40(10):106-109.
[2] 李波,苏经宇,马东辉,等.地震砂土液化判别的灰色关联-逐步分析耦合模型[J].中南大学学报(自然科学版),2016,47(1):232-238.LI Bo,SU Jingyu,MA Donghui,et al.Coupling Model Based on Grey Relational Analysis and Stepwise Discriminant Analysis for Seismic Liquefaction Discrimination of Sandy Soil[J].Journal of Central South University (Science and Technology),2016,47(1):232-238.
[3] 刘年平,王宏图,袁志刚,等.砂土液化预测的Fisher判别模型及应用[J].岩土力学,2012,33(2):554-546.LIU Nianping,WANG Hongtu,YUAN Zhigang,et al.Fisher Discriminant Analysis Model and Its Application to Sand Liquefaction Prediction[J].Hydrogeology & Engineering Geology,2012,33(2):554-546.
[4] 祝百茹,刘海卿.基于模糊综合评判的砂土液化判别[J].广西大学学报(自然科学版),2013,38(1):144-150.ZHU Bairu,LIU Haiqing.Fuzzy Comprehensive Evaluation on Liquefaction of SandSoil[J].Journal of Guangxi University (Nat Sci Ed),2013,38(1):144-150.
[5] 高宗军,付青,郑秋霞,等.BP和Elman神经网络在砂土液化预测中的研究[J].中国安全生产科学技术,2013,9(6):58-62.GAO Zongjun,FU Qing,ZHENG Qiuxia,et al.Study on Forecasting of Sand Liquefaction by Using BP Neural and Elamn Neural Networks[J].Journal of Safety Science and Technology,2013,9(6):58-62.
[6] 李振林.砂土地震液化判别方法的综合应用[J].西安科技大学学报,2010,30(4):451-456.LI Zhenlin.Discrimination Methods for Sandy Soil Seismic Liquefaction[J].Journal of Xi’an University of Science and Technology,2010,30(4):451-456.
[7] 夏建中,罗战友,龚晓南,等.基于支持向量机的砂土液化预测模型[J].岩石力学与工程学报,2005,24(22):4139-4144.XIA Jianzhong,LUO Zhanyou,GONG Xiaonan,et al.Support Vector Machine Model for Predicting Sand Liquefaction[J].Chinese Journal of Rock Mechanics and Engineering,2005,24(22):4139-4144.
[8] 师旭超,郭志涛,韩阳.基于支持向量机的砂土液化预测分析[J].地震工程学报,2009,31(4):363-366.SHI Xuchao,GUO Zhitao,HAN Yang.Analysis on Sand Seismic Liquefaction Prediction Based on the Support Vector Machine[J].China Earthquake Engineering Journal,2009,31(4):363-366.
[9] 张向东,冯胜洋,王长江.基于网格搜索的支持向量机砂土液化预测模型[J].应用力学学报,2011,28(1):24-28.ZHANG Xiangdong,FENG Shengyang,WANG Changjiang.Support Vector Machine Model for Predicting Sand Liquefaction Based on Grid-Search Method[J].Chinese Journal of Applied Mechanics,2011,28(1):24-28.
[10] WANG F,SU J,WANG Z.Forecasting of Building Settlements Due to Earthquake Liquefaction Based on LS-SVM with Mixed Kernel[J].Electronic Journal of Geotechnical Engineering,2015,20(1):11-19.
[11] FAN Z J,LENG Y Q,XU Y L,et al.A Discrimination Method of Saturated Sand Liquefaction Possibility Based on Support Vector Machine[J].Applied Mechanics & Materials,2014,509:38-43.
[12] PAWLAK Z,Rough Set[J].International Journal of Computer and Information Sciences,1982,11:341-356.
[13] 高爽,冬雷,高阳,等.基于粗糙集理论的中长期风速预测[J].中国电机工程学报,2012,32(1):32-37.GAO Shuang,DONG Lei,GAO Yang,et al.Mid-long Term Wind Speed Prediction Based on Rough Set Theory[J].Proceedings of the CSEE,2012,32(1):32-37.
[14] 张文修,吴伟志.粗糙集理论介绍和研究综述[J].模糊系统与数学,2000,14(4):1-12.ZHANG Wenxiu,WU Weizhi.An Introduction and a Survey for the Studies of Rough Set Theory[J].Fuzzy Systems and Mathematics,2000,14(4):1-12.
[15] 张国英,王娜娜,张润生,等.基于主成分分析的BP神经网络在岩性识别中的应用[J].北京石油化工学院学报,2008,16(3):43-46.ZHANG Guoying,WANG Nana,ZHANG Runsheng,et al.Application of Principal Component Analysis and BP Neural Network in Identifying Lithology[J].Journal of Beijing Institute of Petro-chemical Technology,2008,16(3):43-46.
[16] VAPNIK V N.The Nature of Statistical Learning Theory[M].New York:SpringVerlag,1995.
[17] 杨健,陈庆寿.砂土液化影响因素及其判别方法[J].西部探矿工程,2004,93(2):1-2.YANG Jian,CHEN Qingshou.Affecting Factors and Distinguishing Methods of Sandy Soil Liquefaction[J].West-China Exploration Engineering,2004,93(2):1-2.
[18] 蔡煜东,宫家文.砂土液化预测的人工神经网络模型[J].岩土工程学报,1993,15(6):53-58.CAI Yudong,GONG Jiawen.Artificial Neural Network Model for Prediction of Sand Liquefaction[J].Chinese Journal of Geotechnical Engineering,1993,15(6):53-58.
[19] 王举范,陈卓.基于信息熵的粗糙集连续属性多变量离散化算法[J].青岛科技大学学报(自然科学版),2013,34(4):423-426.WANG Jufan,CHEN Zhuo.Multiple Variable Discretization Algorithm of Continuous Attributes in Rough Set Theory Based on Information Entropy[J].Journal of Qingdao University of Science and Technology (Natural Science Edition),2013,34(4):423-426.
[20] WONG S K M,ZIARKO W.On Optiongal Decision Rule in Decision Table[J].Bulletin of Polish Academy of Science,1985,33:693-696.
[21] 郭超,宋卫华,魏威.基于网格搜索-支持向量机的采场顶板稳定性预测[J].中国安全科学学报,2014,24(8):31-36.GUO Chao,SONG Weihua,WEI Wei.Stope Roof Stability Prediction Based on Both SVM and Grid-search Method[J].China Safety Science Journal,2014,24(8):31-36.
[22] 谢玮,王彦春,刘建军,等.基于粒子群优化最小二乘支持向量机的非线性AVO反演[J].石油地球物理勘探,2016,51(6):1187-1194.XIE Wei,WANG Yanchun,LIU Jianjun,et al.Non-linear AVO Inversion Based on PSO-LSSVM[J].Oil Geophysical Prospecting,2016,51(6):1187-1194.
[23] 季斌,周涛发,袁峰.遗传算法优化支持向量机矿产预测方法[J].测绘科学,2015,40(10):106-109.JI Bin,ZHOU Taofa,YUAN Feng.The Detection Method of Maglev Gyroscope Abnormal Data Based on the Characteristics of Two Positioning[J].Science of Surveying and Mapping,2015,40(10):106-109.