地震工程与工程振动
地震工程與工程振動
지진공정여공정진동
EARTHQUAKE ENGINEERING AND ENGINEERING VIBRATION
2010年
1期
53-59
,共7页
Hilbert-Huang变换%随机子空间识别%经验模态分解%自然激励技术%模态参数识别%振动台试验
Hilbert-Huang變換%隨機子空間識彆%經驗模態分解%自然激勵技術%模態參數識彆%振動檯試驗
Hilbert-Huang변환%수궤자공간식별%경험모태분해%자연격려기술%모태삼수식별%진동태시험
Hilbert-Huang transform%stochastic subspace identification%empirical mode decomposition%natural excitation technique%modal parameter identification%shaking table test
基于Hilbert-Huang变换和随机子空间识别技术提出了两种土木工程结构的模态参数识别方法.方法一是基于Hilbert-Huang变换和自然激励技术,通过经验模态分解和Hilbert变换提取信号的瞬时特性.进而利用自然激励技术和模态分析的基本理论识别结构的模态参数;方法二是基于经验模态分解和随机子空间识别技术,通过经验模态分解对信号进行预处理,进而运用随机子空间识别方法处理得到的结构单阶模态响应以识别结构的模态参数.利用这两种方法,通过对一12层钢筋混凝土框架模型振动台试验测点加速度记录的处理,识别了该模型结构的模态参数.识别结果与传统的基于傅里叶变换的识别结果及有限元分析结果的对比验证了这两种方法的可行性和实用性.
基于Hilbert-Huang變換和隨機子空間識彆技術提齣瞭兩種土木工程結構的模態參數識彆方法.方法一是基于Hilbert-Huang變換和自然激勵技術,通過經驗模態分解和Hilbert變換提取信號的瞬時特性.進而利用自然激勵技術和模態分析的基本理論識彆結構的模態參數;方法二是基于經驗模態分解和隨機子空間識彆技術,通過經驗模態分解對信號進行預處理,進而運用隨機子空間識彆方法處理得到的結構單階模態響應以識彆結構的模態參數.利用這兩種方法,通過對一12層鋼觔混凝土框架模型振動檯試驗測點加速度記錄的處理,識彆瞭該模型結構的模態參數.識彆結果與傳統的基于傅裏葉變換的識彆結果及有限元分析結果的對比驗證瞭這兩種方法的可行性和實用性.
기우Hilbert-Huang변환화수궤자공간식별기술제출료량충토목공정결구적모태삼수식별방법.방법일시기우Hilbert-Huang변환화자연격려기술,통과경험모태분해화Hilbert변환제취신호적순시특성.진이이용자연격려기술화모태분석적기본이론식별결구적모태삼수;방법이시기우경험모태분해화수궤자공간식별기술,통과경험모태분해대신호진행예처리,진이운용수궤자공간식별방법처리득도적결구단계모태향응이식별결구적모태삼수.이용저량충방법,통과대일12층강근혼응토광가모형진동태시험측점가속도기록적처리,식별료해모형결구적모태삼수.식별결과여전통적기우부리협변환적식별결과급유한원분석결과적대비험증료저량충방법적가행성화실용성.
Two approaches are proposed for modal parameter identification of civil engineering structures based on Hilbert-Huang transform (HHT) and stochastic subspace identification (SSI). The first approach is based on HHT and natural excitation technique (NExT). Empirical mode decomposition (EMD) and Hilbert transform (HT) are used to extract the instantaneous characteristics of the original signal. Then, NExT and basic modal analysis theory are applied to identify modal parameters. The second approach is based on EMD and SSI. Each single modal response is obtained through processing the original signal by EMD, then the modal parameters are identified by SSI. Finally, the original signals from the shaking table test of a 12-storey reinforced concrete frame model are processed and modal parameters are identified by these approaches, respectively. Identification results and comparison with the results from traditional fast Fourier transform (FFT) and finite element analysis indicate that the proposed approaches are feasible and practical.
