地震砂土液化判别的灰色关联-逐步分析耦合模型

来源期刊:中南大学学报(自然科学版)2016年第1期

论文作者:李波 苏经宇 马东辉 王威

文章页码:232 - 239

关键词:砂土液化;灰色关联分析;逐步判别分析;耦合判别模型

Key words:liquefaction of sandy soil; grey relational analysis; stepwise discriminant analysis; coupling discriminant model

摘    要:通过分析灰色关联模型和逐步判别模型在单独进行砂土液化等级判别时所体现的优点及存在的问题,建立以两者为基础的耦合判别模型。利用实际样本结果,选取震级M、地面加速度最大值gmax、比贯入阻力Ps、标准贯入击数N63.5、平均粒径D50、相对密实度Dr、地下水位dw共7个实测数据作为砂土液化判别因子,对该模型进行验证。研究结果表明:BP神经网络、灰色关联分析、逐步判别分析这3种方法对样本的判别准确率分别为60%,80%和60%,耦合模型的判别准确率为100%。与单独使用2种基础模型相比,耦合模型的判别结果与实际结果更加吻合,表明该方法具有较高的准确性和良好的实用性。

Abstract: The advantages and disadvantages of grey relational model and stepwise discriminant model were analyzed when both of them were used alone for evaluation of seismic liquefaction of sandy soil, and coupling model was established based on both models. Seven sets of factors (magnitude M, the maximum ground acceleration gmax, specific penetration resistance Ps, standard penetration blow count N63.5, average particle size of D50, relative density Dr, water table dw) were selected for model verification. The results show that the accuracy of BP neural network model, grey relational model and stepwise discriminant model are 60%, 80% and 60%, respectively. The accuracy of coupling model is 100%, it indicates that the discriminant results are more compatible with the actual results compared with one analysis method alone, and that the coupling model has high accuracy and good practicability.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号