基于样本点选择策略的一种改进响应面法  被引量:3

An improved response surface method based on sample point selection strategy

在线阅读下载全文

作  者:夏雨[1] 张泽俊 余颖烨[1] 刘敬敏 Xia Yu;Zhang Zejun;Yu Yingye;Liu Jingmin(School of Architecture and Civil Engineering,Guangxi University of Science and Technology,545006,Liuzhou,China)

机构地区:[1]广西科技大学土木建筑工程学院,柳州545006

出  处:《应用力学学报》2021年第1期176-182,共7页Chinese Journal of Applied Mechanics

基  金:国家自然科学基金(51569005);广西自然科学基金(2020JJA160008,2018GXNSFBA281186,2019GXNSFBA245071,桂科AD19110068,桂科AD19245125);广西高校中青年教师基础能力提升项目(KY2016YB258,2018KY0318)。

摘  要:为提高响应面函数在验算点附近的拟合精度,提出了一种基于样本点选择策略的改进响应面法,选取不考虑随机变量耦合项的多项式进行拟合,通过对第二次迭代产生的设计点构造参考点,令样本点或参考点进行线性插值,从而得到下一次迭代所需样本点。该方法不仅能有效利用已有的抽样信息从而减少评估结构系统可靠性所需的计算工作量,还可以使得拟合得到的响应面更好地呈现验算点附近极限状态面的非线性趋势,从而提高失效概率的评估精度。算例表明:该方法在进行隐式或显示极限状态函数下的可靠度计算中,相对传统响应面法均提高了一定的效率和精度,具有一定的工程实际意义。In order to improve the fitting accuracy of the response surface function near the test point, an improved response surface method based on the sample point selection strategy is proposed in this paper. The polynomial is selected without considering the coupling terms of random variables for fitting. A reference point is constructed and linear interpolation on the sample points or reference points is performed to obtain the sample points required for the next iteration. This method can not only effectively use the existing sampling information to reduce the calculation workload required to evaluate the reliability of the structural system, but also make the fitted response surface better present the nonlinear trend of the limit state surface near the check point, thereby improving the assessment accuracy of failure probability. Numerical examples show that the method is of higher efficiency and accuracy under implicit or explicit limit state functions, and of certain engineering practical significance.

关 键 词:结构可靠度 响应面法 样本点 失效概率 

分 类 号:TU[建筑科学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象