應用時頻分解法強化鋼筋混凝土結構地震反應之解析

鋼筋混凝土結構受到較大的作用力時，如地震的影響，致使結構進入非線性或非彈性行為。因為地震工程上主要考慮結構受到地震作用後的行為狀態與損傷情形。根據過去文獻報告中，發現使用較複雜的非線性模型模擬RC結構之非線性行為，如Bouc-Wen 模型等。也因為Bouc-Wen 模型需要識別的較多的未知參數與計算複雜，因此為了減少未知參數來模擬RC之非線性行為，因此本文研究議題則使用較簡單與參數較少的損耗分佈元素模型(Deteriorating Distributed Element model) 來模擬RC結構之地震反應。本文主要研究目的是應用損耗分佈元素模型(DDE model)，來模擬鋼筋混凝土結構之非線性反應。並依照訊號處理的方法，由結構的遲滯迴圈識別DDE模型的參數。首先應用線性奇異譜分析(SSA)將RC反應之穩態訊號萃取出來；使用非線性奇異譜分析(NLSSA)法分析非穩態訊號，以萃取單頻穩定的訊號。最後應用DDE 模型模擬RC結構非線性行為與損傷情形，且由分解成許多為單頻的訊號之絕對加速度與位移，及其對應的遲滯迴圈，估計實際結構的有效勁度與對等阻尼比，由此再計算DDE模型之有效勁度參數。從分析結果發現損耗分佈元素模型(DDE model)模擬RC結構非線性行為，其時間歷時之絕對加速度與位移等和實際RC結構反應比較，都能互相對應。本文最後以解析度較高的遞迴隨機子空間識別法(RSSI)驗證損耗分佈元素模型(DDE model)模擬RC結構之非線性反應或行為的可靠性。

關鍵字

損耗分佈元素模型；奇異譜分析；非線性奇異譜分析；訊號處理；有效勁度；對等阻尼比；遞迴隨機子空間識別法

並列摘要

Reinforced concrete structures typically exhibit nonlinear and inelastic behavior under severe dynamic loading, such as under earthquakes. Earthquake engineering considers both the expected performance of structures in an earthquake and the state of the damage after seismic events. According to the literature review, it is clear that using complex non-linear model to simulate the nonlinear behavior of the RC structures, such as the Bouc-Wen model, may need to identify many unknown model parameters. To reduce the number of unknow parameters in modeling the nonlinear and inelastic behavior of the seismic response of reinfoorced concrete, the Deteriorating Distributed Element model was used in this study to simulate the seismic response of the RC structures. The objective of this research is to apply the Deteriorating Distributed Element model to simulate the nonlinear response of reinforced concrete structures. An advanced signal processing technique is used to investigate the physical parameters of a inelastic system in the degrading hysteretic model of a reinforced concrete frame. First, the Singular Spectrum Analysis was used to extract the steady-state signals from the nonlinear RC signal; and then the Nonlinear Singular Spectrum Analysis method was adapt to analyze the non-stationary signal of seismic response of RC frame so as to extract a stable single-frequency signals. Through the singular spectrum analysis (SSA) and the nonlinear SSA, the original seismic response data is decomposed into independent additive components of decreasing weight which can be used for developing hysteresis loop also in degrading order. Finally, the Deteriorating Distributed Element (DDE) model is applied to simulate the damage pattern of the reinforced concrete frame with respect to each of the developed degrading hysteresis loop and to identify the degrading stiffness and equivalent damping ratio. Comparison between the developed DDE model with the recorded physical indices and the results of structural system identification is also discussed. Finally, the Recursive Stochastic Subspace Identification method was applied, to verify the nonlinear response of the RC structural reliability through the simulation of using the DDE model.

並列關鍵字

Deteriorating Distributed Element model ； Singular Spectrum Analysis ； Nonlinear Singular Spectrum Analysis ； signal processing ； the degrading stiffness ； equivalent damping ratio ； Recursive Stochastic Subspace Identification

參考文獻

1. Hsieh, W.W. Nonlinear multivariate and time series analysis by neural network

2. Giraldo DF, Dyke SJ, Caicedo JM. Damage detection accommodating varying environmental conditions. Structural Health Monitoring 2006; 5:155–172.

3. Yan AM, Kerschen G, De Boe P, Golinval JC. Structural damage diagnosis under changing environmental conditions-part I: a linear analysis. Mechanical Systems and Signal Processing, 2005; 19: 847–864.

4. Hsieh, W.W. , & Wu A. Nonlinear multichannel singular spectrum analysis of the tropical Pacific climate variability using a neural network approach. Journal of Geophysical Research (Oceans), 2001, 1-36.

5. Hsieh, W.W., & Wu A. Nonlinear singular spectrum analysis by neural networks. Neural Networks, 2001, 1-34.

國際替代計量

應用時頻分解法強化鋼筋混凝土結構地震反應之解析

全文下載

主題瀏覽