引用楼主civilwyf于2007-06-27 15:42发表的 从新加坡尼诰大道塌陷谈谈PLAXIS的缺陷！ :
临时墙壁倒塌，使支撑尼诰大道地铁站附近的地下工地结构如骨牌倒塌，连同尼诰大道（Nicoll Highway)的一大片路面也下陷30公尺，双程方向交通全面中断。这起历来最严重的地铁工程意外造成1人死，3人轻伤，另3人失踪。民防部队出动百多名拯救人员，连同搜救犬搜寻和抢救受困地下工地废墟的3个人。到截稿为止，还没有他们的下落，但搜寻工作将彻夜进行。
坍塌路面100公尺长，150公尺宽，有两个足球场大。马路东面的地铁站工地也受波及，地面崩陷。路面坍塌地点是黄金地带购物中心后面的路段，距离独立桥（Merdeka Bridge）大约半公里，离拟建中的尼诰大道地铁站约250公尺。那里正在进行地铁环线的工事。
该项目中使用了plaxis软件计算其深开挖的最大侧向唯一变形，用了elastic and perfectly plastic Mohr-Coulomb model,参数采用了有效应力参数。 事故后发现实际侧向位移变形被plaxis计算严重低估，这被视作是事故的原因之一。
.......
I'd a thought that there have many factors lead to the failure of Nicole highway, obviously Geotechnical FE analysis is just one of those.
In Geo FEM analysis, it is seldom using effective parameters to model undrain behaviour. but PLAXIS does show a example in the tutorial, and new Plaxis user would be easily to missunderstand if they are not really understand the soil behaviour.
it would be clear if you look at the stress path.for MC model for low excavation in clay, it is conservative and it is safe. but if going to deep excavation in clay, using effective parameters to model undrained behaviour would underestimate the porepressure and overestimate the undrained shear strength, which was adopted by the designer in Nicole highway project.
it should be known that in Plaxis, in undrained condition, we have to input the correct Cu value rather than generated by Plaxis, which is laterly call Method B.
welcom to disscuss, Feel free to drop me an email.
Cheers, yk
