• 欢迎访问ERC!
引用本文:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
过刊浏览    高级检索
本文已被:浏览 76次   下载 0  
分享到: 微信 更多
基于贝叶斯正则化神经网络的粉质粘土压缩性参数取值模拟
蔡润,彭涛,王谦,何蕃民,赵多银
1.中冶成都勘察研究总院有限公司;2.中国地震局甘肃省 黄土地震工程重点实验室
摘要:
土体压缩性参数是岩土领域重要的参数指标,受自然条件、人为干扰等多种因素的影响。当样本数量过大时,常规方法测定压缩性参数往往需要大量人力物力。为合理模拟样本的压缩性参数,首先结合相关性分析,挑选出与压缩性参数相关性较高的五个影响因素,同时利用Garson理论对五个参数在贝叶斯神经网络的权值占比进行分析。其次基于贝叶斯正则化BP神经网络建立BR-BP粉质粘土压缩性参数的输出模型。最后利用该模型对某工程实测的压缩性参数进行仿真试验,并将输出结果同实测值和传统LM-BP神经网络的输出结果进行对比,结果表明本文所建模型更加稳定,具有更强的非线性拟合能力,外推能力更强。模型输出结果与实际值基本吻合,与传统的LM-BP神经网络模型相比,其数据敏感性更强,输出结果精度得到显著提高,为该地区粉质粘土的压缩性参数获取提供了一种新的技术方法和手段,具有较好的理论意义和实践价值。
关键词:  粉质粘土  压缩性  相关性分析  贝叶斯正则化  神经网络
DOI:
分类号:
基金项目:中国地震局地震预测研究所基本科研业务经费项目(2018IESLZ06),国家自然科学(51778590)和地震科技星火计划项目(XH20057)共同资助.
Simulation of compressibility parameters of silty clay Based onImproved BP Neural Network Using Bayesian Regularization
CAI Run1,2,3,2,4,2, PENG Tao1, WANG Qian5, HE Fanmin1, ZHAO Duoying1
1.Chengdu Surveying Geotechnical Research Institute Co Ltd of MCC;2.China;3.Key Laboratory of Loess Earthquake Engineering,China Earthquake Administration and Gansu Province,Lanzhou,;4.Lanzhou Institute of Seismology,China Earthquake Administration,Lanzhou,;5.Key Laboratory of Loess Earthquake Engineering,China Earthquake Administration and Gansu Province,Lanzhou
Abstract:
Soil compressibility parameters are important parameter indicators in the geotechnical field, they are affected by various factors such as natural conditions and human factors. When the sample size is too large, Conventional methods to measure compressibility parameters often require a lot of human and financial resources. In order to reasonably simulate the compressibility parameters of the sample. Firstly, this paper adopts the correlation analysis to select seven influencing factors that have a high correlation with the compressibility parameters, at the same time, the proportion of the weights of the seven parameters in the Bayesian neural network is analyzed by using Garson theory. Secondly, an output model of compressibility parameters of BR-BP silty clay is established based on Bayesian regularized BP neural network. Finally, the model is used to simulate the measured compressibility parameters of a project, and the output results are compared with the measured values and the output results of the traditional LM-BP neural network. The results show that the model is more stable and has stronger nonlinear fitting ability and stronger extrapolation. The output of the model is basically consistent with the actual value. Compared with the traditional LM-BP neural network model, its data sensitivity is enhanced, and the accuracy of the output result is significantly improved. It provides a new technical method and means for obtaining compressibility parameters of silty clay in this area, which has good theoretical significance and practical value.
Key words:  silty clay  compressibility  Correlation analysis  Bayesian regularization  Neural network