基于VMD和DBN的非线性结构模型参数识别  被引量:1

Parametric recognition of nonlinear structural model based on VMD and DBN

在线阅读下载全文

作  者:莫叶 王佐才[1,2] 丁雅杰 袁子青 MO Ye;WANG Zuocai;DING Yajie;YUAN Ziqing(School of Civil Engineering,Hefei University of Technology,Hefei 230009,China;Anhui Engineering Research Center for Civil Engineering Disaster Prevention and Mitigation,Hefei,230009,China)

机构地区:[1]合肥工业大学土木与水利工程学院,合肥230009 [2]安徽省土木工程防灾减灾工程技术研究中心,合肥230009

出  处:《振动与冲击》2022年第9期136-143,共8页Journal of Vibration and Shock

基  金:国家自然科学基金优秀青年科学基金(51922036);安徽省重点研发计划(1804a0802204);中央高校基本科研业务费专项资金(JZ2020HGPB0117)。

摘  要:为解决现有的非线性结构模型参数识别方法面临优化过程复杂的问题,提出一种基于变分模态分解(variational mode decomposition,VMD)和深度置信网络(deep belief network,DBN)的非线性结构模型参数识别方法。首先,利用VMD和希尔伯特变换(Hilbert transform,HT)识别振动响应的瞬时参数;将瞬时参数进行主成分分析后作为输入,非线性模型参数作为输出;然后,利用DBN拟合两者之间的非线性映射关系;最后,将实测振动响应的瞬时参数进行主成分分析,输入训练好的DBN可直接识别修正后的非线性模型参数。通过对两个不同非线性类型的双自由度模型和一个复杂框架模型在地震作用下的数值模拟,与高压输电结构的振动台试验,验证了该方法的有效性。数值与试验结果表明,所提方法具有较高的计算效率和良好的抗噪性。Here,to solve the problem of existing nonlinear structural model parametric recognition methods facing complex optimization process,a parametric recognition method of nonlinear structural model based on variational mode decomposition(VMD)and deep belief network(DBN)was proposed.Firstly,VMD and Hilbert transform(HT)were used to recognize instantaneous parameters of vibration response,instantaneous parameters after principal component analysis were taken as the input,and the nonlinear model parameters were taken as the output.Then,DBN was used to fit the nonlinear mapping relation between them.Finally,the measured vibration response’s instantaneous parameters after principal component analysis were input into the well-trained DBN to directly identify the modified nonlinear model parameters.The effectiveness of the proposed method was verified with numerical simulations of two different nonlinear 2 DOF models and a complex frame model under earthquake,and shaking table tests of high-voltage transmission structure.Numerical and test results showed that the proposed method has higher computational efficiency and good noise immunity.

关 键 词:非线性结构模型 参数识别 变分模态分解(VMD) 深度置信网络(DBN) 振动响应 瞬时参数 

分 类 号:O322[理学—一般力学与力学基础]

 

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