砂土液化预测的Fisher判别分析模型及应用
|
摘要
针对砂土地震液化预测问题,基于Fisher判别分析原理(FDA),选用平均粒径、不均匀系数、标贯击数、地下水水位、砂层埋深、剪应力与有效上覆应力比、地震烈度、震中距等8个实测指标为判别因子,建立Fisher线性判别函数模型,对砂土液化进行预测。研究结果表明,Fisher判别分析结果与神经网络输出结果一致,优于规范法和Seed法判别结果,验证了该模型的合理性和可靠性。运用该模型进行判别分析,简易方便,分类效率高,对砂土液化判别快速、有效,模型适用性强,具有一定的工程应用前景。
For the prediction of seismic liquefaction of sand,Fisher discriminant model is established to the prediction of sand liquefaction based on the principle of Fisher discriminant analysis theory with eight factors listed as follows: mean diameter,coefficient of non-uniformity,blow number of standard penetration test,underground water depth,sand depth,ratio of shearing stress to effective overburden stress,seismic intensity,epicenteral distance.Through computing practical examples and assessing the model,the Fisher discriminant analysis model is manifested to be of rationality and reliability and the model identification results are consistent with the neural network output,and are better than the discriminant results with the specification method and Seed method.It is easy and efficient to make discriminant analysis using this model and it can provide efficient means to predict sand liquefaction.
For the prediction of seismic liquefaction of sand,Fisher discriminant model is established to the prediction of sand liquefaction based on the principle of Fisher discriminant analysis theory with eight factors listed as follows: mean diameter,coefficient of non-uniformity,blow number of standard penetration test,underground water depth,sand depth,ratio of shearing stress to effective overburden stress,seismic intensity,epicenteral distance.Through computing practical examples and assessing the model,the Fisher discriminant analysis model is manifested to be of rationality and reliability and the model identification results are consistent with the neural network output,and are better than the discriminant results with the specification method and Seed method.It is easy and efficient to make discriminant analysis using this model and it can provide efficient means to predict sand liquefaction.
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 D Z,YUAN J Y.Soil science and the soil mechanics[M].Beijing:Communications Press,2001:216-222(inChinese)]
[2]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):451-455.[LIZ L.Discrimination methods for sandy soil seismicliquefaction[J].Journal of Xi’an University ofScience and Technology,2010,30(4):451-455.(in Chinese)]
[3]张菊连,沈明荣.基于逐步判别分析的砂土液化预测研究[J].岩土力学,2010,31(增1):298-304.[ZHANG J L,SHEN M R.Sand liquefactionprediction based on stepwise discriminant analysis[J].Rock and Soil Mechanics,2010,31(S1):298-304.(in Chinese)]
[4]黄志全,姜彤,李玉起.基于组合权重的模糊物元模型在砂土液化势评价中的应用[J].世界地震工程,2009,25(4):126-131.[HUANG Z Q,JIANGT,LI Y Q.Application of fuzzy matter-elementmodel based on coefficients of combined weights inevaluation of sand liquefaction[J].World EarthquakeEngineering,2009,25(4):126-131.(inChinese)]
[5]罗战友,龚晓南.基于经验的砂土液化灰色关联系统分析与评价[J].工业建筑,2002,32(11):36-39.[LUO Z Y,GONG X N.Grey incidencesystem analysis and evaluation for liquefaction of sandsbased on experience[J].Industrial Construction,2002,32(11):36-39.(in Chinese)]
[6]马骥,傅光翮,罗国煜.基于MATLAB的BP神经网络在砂土液化评价中的应用[J].水文地质工程地质,2004,31(2):54-58.[MA J,FU G H,LUO G Y.Application of BP neural network based onMATLAB to the evaluation of sandy soil seismicliquefaction[J].Hydrogeology&EngineeringGeology,2004,31(2):54-58.(in Chinese)]
[7]陈国兴,李方明.基于径向基函数神经网络模型的砂土液化概率判别方法[J].岩土工程学报,2006,28(3):301-305.[CHEN G X,LI F M.Probabilistic estimation of sand liquefaction based onneural network model of radial basis function[J].Chinese Journal of Geotechnical Engineering,2006,28(3):301-305.(in Chinese)]
[8]王星华,崔科宇,周海林.振动力作用下饱和砂土液化的模糊神经网络预测[J].中国铁道科学,2010,31(6):26-31.[WANG X H,CUI K Y,ZHOU H L,et al.Fuzzy neural network forecast forthe liquefaction of the saturated sand under the actionof the vibration force[J].China Railway Science,2010,31(6):26-31.(in Chinese)]
[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[J].应用力学学报,2011,28(1):24-28.[ZHANG X D,FENG S Y,WANGC J.Support vector machine model for predicting sandliquefaction based on Grid-Search method[J].Chinese Journal of Applied Mechanics,2011,28(1):24-28.(in Chinese)]
[10]金志仁.基于距离判别分析方法的砂土液化预测模型及应用[J].岩土工程学报,2008,30(5):776-780.[JIN Z R.Prediction of sand liquefactionbased on distance discriminant analysis and itsapplication[J].Chinese Journal of GeotechnicalEngineering,2008,30(5):776-780.(inChinese)]
[11]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 J,SHI X Z.Fisher discriminantanalysis method for prediction of classification of rockburst risk[J].Journal of China Coal Society,2010,35(S1):22-27.(in Chinese)]
[12]袁志刚,王宏图,胡国忠,等.煤层注水难易程度的Fisher判别分析模型及应用[J].煤炭学报,2011,36(4):638-642.[YUAN Z G,WANG H T,HU G Z,et al.Fisher discriminant analysis modeland application of classifying difficulty degree of waterinfusion for coal seam[J].Journal of China CoalSociety,2011,36(4):638-642.(in Chinese)]
[13]费宇.应用数理统计-基本概念与方法[M].北京:科学出版社,2007:155-157.[FEI Y.Application of mathematical statistics-basic conceptsand methods[M].Beijing:Science PublishingCompany,2007:155-157.(in Chinese)]
[14]范金城,梅长林.数据分析[M].北京:科学出版社,2002:147-148.[FAN J C,MEI C L.DataAnalysis[M].Beijing:Science PublishingCompany,2002:147-148.(in Chinese)]
[2]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):451-455.[LIZ L.Discrimination methods for sandy soil seismicliquefaction[J].Journal of Xi’an University ofScience and Technology,2010,30(4):451-455.(in Chinese)]
[3]张菊连,沈明荣.基于逐步判别分析的砂土液化预测研究[J].岩土力学,2010,31(增1):298-304.[ZHANG J L,SHEN M R.Sand liquefactionprediction based on stepwise discriminant analysis[J].Rock and Soil Mechanics,2010,31(S1):298-304.(in Chinese)]
[4]黄志全,姜彤,李玉起.基于组合权重的模糊物元模型在砂土液化势评价中的应用[J].世界地震工程,2009,25(4):126-131.[HUANG Z Q,JIANGT,LI Y Q.Application of fuzzy matter-elementmodel based on coefficients of combined weights inevaluation of sand liquefaction[J].World EarthquakeEngineering,2009,25(4):126-131.(inChinese)]
[5]罗战友,龚晓南.基于经验的砂土液化灰色关联系统分析与评价[J].工业建筑,2002,32(11):36-39.[LUO Z Y,GONG X N.Grey incidencesystem analysis and evaluation for liquefaction of sandsbased on experience[J].Industrial Construction,2002,32(11):36-39.(in Chinese)]
[6]马骥,傅光翮,罗国煜.基于MATLAB的BP神经网络在砂土液化评价中的应用[J].水文地质工程地质,2004,31(2):54-58.[MA J,FU G H,LUO G Y.Application of BP neural network based onMATLAB to the evaluation of sandy soil seismicliquefaction[J].Hydrogeology&EngineeringGeology,2004,31(2):54-58.(in Chinese)]
[7]陈国兴,李方明.基于径向基函数神经网络模型的砂土液化概率判别方法[J].岩土工程学报,2006,28(3):301-305.[CHEN G X,LI F M.Probabilistic estimation of sand liquefaction based onneural network model of radial basis function[J].Chinese Journal of Geotechnical Engineering,2006,28(3):301-305.(in Chinese)]
[8]王星华,崔科宇,周海林.振动力作用下饱和砂土液化的模糊神经网络预测[J].中国铁道科学,2010,31(6):26-31.[WANG X H,CUI K Y,ZHOU H L,et al.Fuzzy neural network forecast forthe liquefaction of the saturated sand under the actionof the vibration force[J].China Railway Science,2010,31(6):26-31.(in Chinese)]
[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[J].应用力学学报,2011,28(1):24-28.[ZHANG X D,FENG S Y,WANGC J.Support vector machine model for predicting sandliquefaction based on Grid-Search method[J].Chinese Journal of Applied Mechanics,2011,28(1):24-28.(in Chinese)]
[10]金志仁.基于距离判别分析方法的砂土液化预测模型及应用[J].岩土工程学报,2008,30(5):776-780.[JIN Z R.Prediction of sand liquefactionbased on distance discriminant analysis and itsapplication[J].Chinese Journal of GeotechnicalEngineering,2008,30(5):776-780.(inChinese)]
[11]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 J,SHI X Z.Fisher discriminantanalysis method for prediction of classification of rockburst risk[J].Journal of China Coal Society,2010,35(S1):22-27.(in Chinese)]
[12]袁志刚,王宏图,胡国忠,等.煤层注水难易程度的Fisher判别分析模型及应用[J].煤炭学报,2011,36(4):638-642.[YUAN Z G,WANG H T,HU G Z,et al.Fisher discriminant analysis modeland application of classifying difficulty degree of waterinfusion for coal seam[J].Journal of China CoalSociety,2011,36(4):638-642.(in Chinese)]
[13]费宇.应用数理统计-基本概念与方法[M].北京:科学出版社,2007:155-157.[FEI Y.Application of mathematical statistics-basic conceptsand methods[M].Beijing:Science PublishingCompany,2007:155-157.(in Chinese)]
[14]范金城,梅长林.数据分析[M].北京:科学出版社,2002:147-148.[FAN J C,MEI C L.DataAnalysis[M].Beijing:Science PublishingCompany,2002:147-148.(in Chinese)]
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心