基于紧支径向基响应面的承载力不确定性分析

来源期刊:中南大学学报(自然科学版)2014年第10期

论文作者:胡常福 任伟新 刘旭政

文章页码:3621 - 3629

关键词:极限承载力;不确定性分析;紧支径向基函数;增广基函数

Key words:ultimate load-carrying capacity; uncertainty analysis; compactly supported radial basis functions; augmented functions

摘    要:针对极限承载力不确定性分析中蒙特卡洛有限元计算规模过大的问题,引入增广紧支径向基响应面作为参数与响应的近似替代函数,以解决蒙特卡洛有限元的计算成本问题。以4个典型荷载工况下钢管混凝土肋拱极限承载力不确定性分析为研究对象,分别在响应面拟合精度与蒙特卡洛模拟结果精度方面对增广紧支径向基响应面与传统二次多项式响应面进行比较,并对不同径向基函数、增广基函数和试验设计方法对增广紧支径向基响应面蒙特卡洛模拟结果的影响进行分析。研究结果表明:增广紧支径向基响应面可以用于钢管混凝土拱极限承载力不确定分析中,试验次数少且精度高;增广紧支径向基响应面的拟合精度和蒙特卡洛模拟精度均比传统二次多项式响应面的拟合精度高,不同紧支径向基函数形式对结果影响不大;使用线性增广基紧支径向基响应面计算极限承载力不确定性的均值与标准差最大相对误差均小于1%,在均匀设计与中心复合试验设计中能兼顾响应面精度与试验成本。

Abstract: For too large computation scale of Monte Carlo finite element method used in ultimate load-carrying capacity uncertainty analysis problem, response surface method with compactly supported radial basis functions was used as surrogate models of response and parameters to solve this problem. In four typical load cases, uncertainty problems of concrete filled steel tubular arch load carrying capacity were taken as examples, response surface method with augmented compactly supported radial basis functions was compared with traditional quadratic polynomial response surface method in response surface fitting accuracy and Monte Carlo simulation accuracy, and the influence of different radial basis functions, augmented reponse surface and design of experiment in load carrying capacity uncertainty problems were analyzed. The results show that the functions method with the augmented compactly supported radial basis has high precision and small sample points in load carrying capacity uncertainty problems, and it has higher response surface fitting accuracy and Monte Carlo simulation accuracy than those of the quadratic polynomial response surface method, and different compactly radial basis functions have no influence in this problems, the relative error of mean and standard deviation of linear polynomial augmented compactly supported radial basis functions is less than 1% compared with that of Monte Carlo finite element results, and uniform design and central composite design can take the experiment cost and result accuracy into account.

相关论文

  • 暂无!

相关知识点

  • 暂无!

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