結合識別技術與嵌入式統計模式之損害診斷

Reference-Based Damage Diagnosis through Identification Technique and Embedded Statistical Model

吳艾倫(Ai-Lun Wu)

指導教授 ： 羅俊雄

國立臺灣大學/工學院/土木工程學研究所/碩士(2006年)

https://doi.org/10.6342/NTU.2006.00073

摘要

結構系統的模態參數如自然頻率、阻尼等會隨著系統產生破壞而有所改變，然而藉由識別模態參數只能進一步偵測系統的損壞，無法量化損害程度。本文提出一套損害診斷的方法，針對結構系統，不僅以損害指標量化損害情形，更深入探討系統剛度及強度的折減情形，對系統的損害做更深入的解釋。因此，配合識別技巧與嵌入式統計模式，可提供系統受地震力作用之下，剛度與強度折減程度，配合損害指標，更完整描述系統的損害情形。
嵌入式統計模式主要是以一雙線性模式為參考基底，考慮不同遲滯行為效應，經由統計分析而建立。針對正規化遲滯能以及考慮系統週期、最大位移之γ-spectrum，以及系統地震力折減係數分別建立其嵌入式統計模式。此統計模式包含了雙線性的參考基底，以及考慮剛度折減，強度衰弱及緊縮行為的校正因子.
一方面藉由建置完成的嵌入式統計模式，可供預測γ-spectrum 與系統地震力折減係數，另一方面，藉由量測結構系統各層反應，利用識別技巧，由遲滯迴圈可識別遲滯模型的參數。配合正規化遲滯能之嵌入式統計模式，可求得系統剛度折減與強度衰弱的程度，配合損害指標，量化系統整體損害情形。最後以數值模擬及實例針對鋼筋混凝土構架進行驗證，結果顯示此套評估方法可深入探討系統不同遲滯行為對損害的影響程度。

關鍵字

系統識別

並列摘要

Structural damage will cause the change of modal characteristics, such as natural frequencies, mode shapes or modal damping. However these dynamic characteristics can not quantify the damage of the structure. The purpose of this paper is to develop a method for damage diagnosis of a sub-structural system and identify the degree of damage which can be interpreted not only on the damage index but also on the percentage of strength and stiffness degradation. Based on parameter identification of component inelastic hysteretic behavior and the proposed statistical damage model, the damage index and the degree of structural damage can be identified.
To develop the statistical damage model a reference inelastic hysteretic behavior (a generalized bi-linear model) is assumed. Three statistical models of response indices are developed: Normalized hysteretic energy (NHE) ,γ-spectrum and reduction factor, using simulation of seismic response data. These models contain the calibration and modification factors of the post-yielding stiffness, the strength and stiffness degradation, and the pinching effects. These models are obtained from statistical analysis.
For constructed statistical model of γ-spectrum and reduction factor , one can used for prediction. On the other hand, based on the measurement of floor structural responses, particularly the structural component restoring force diagram of floor system, the model parameters of the inelastic hysteretic model is identified first. Using the proposed statistical model of normalized hysteretic energy, the damage index as well as the percentage of stiffness and strength degradation can be quantified
This proposed model is verified both numerically and experimentally using RC frame structures. The results indicate that the current approach can quantify the degree of damage of RC frame structure.

並列關鍵字

system identification

參考文獻

H. Banon and D. Veneziano. (1982). ‘‘Seismic safety of reinforced concrete members and structures.’’ Earthquake Eng. Struct. Dyn., Vol.6, pp. 31-42.

E. Cozenza, G. Manfredi, and R. Ramasco. (1993). ‘‘The use of damage functionals in earthquake engineering: A comparison between different method. ’’ Earthquake Eng. Struct. Dyn., Vol.22, pp855-868.

Y. H. Chai, Romstad, K. M., and Bird, S. M. (1995). ‘‘Energy-based linear damage model for high-intensity seismic loading.’’ J. Struct. Eng., 121(5), 857–864.

C. H. Chen (2005). “Structural identification from field measurement data using a neural networks.” Smart Mater. Struct. 14, S104-S115.

S. H. Chao and C. C. Loh. (2005). “Dynamic response analysis of frame structures using forced analogy method with considering deteriorating inelastic hysteretic model & P- effect .” Earthquake Engineering and Structural Dynamics.