【摘要】：地震是破壞力最強(qiáng)的一種自然災(zāi)害之一。精確再現(xiàn)建筑結(jié)構(gòu)的地震響應(yīng)對(duì)于提高結(jié)構(gòu)的抗震能力，防止建筑結(jié)構(gòu)在地震中倒塌具有重要的意義。常規(guī)再現(xiàn)結(jié)構(gòu)地震響應(yīng)的方法包括數(shù)值模擬和物理實(shí)驗(yàn)，這兩種方法通常能得到較為精確的結(jié)果，但是也都有一定的局限性�；旌显诰�實(shí)驗(yàn)技術(shù)能夠充分利用數(shù)值模擬和物理實(shí)驗(yàn)的優(yōu)勢(shì)，即：利用數(shù)值模擬技術(shù)建立結(jié)構(gòu)的運(yùn)動(dòng)平衡方程，而通過物理實(shí)驗(yàn)獲得結(jié)構(gòu)的恢復(fù)力特性，可以精確有效地模擬結(jié)構(gòu)的地震響應(yīng)，是一種非常有前途的技術(shù)。本研究開發(fā)了一種結(jié)合有限元軟件和子結(jié)構(gòu)技術(shù)的新型混合在線實(shí)驗(yàn)系統(tǒng),本論文重點(diǎn)關(guān)注如何改進(jìn)在線混合實(shí)驗(yàn)中數(shù)值子結(jié)構(gòu)的模擬精度。系統(tǒng)將用戶子程序接口VDISP作為主控程序和ABAQUS/Explicit模塊的接口，通過此結(jié)構(gòu)向ABAQUS/Explicit模塊輸入目標(biāo)位移并得到反作用力。文中之所以采用ABAQUS/Explicit模塊，，而不是ABAQUS/Standard模塊是因?yàn)椋瑥?fù)雜結(jié)構(gòu)容易出現(xiàn)收斂性問題，而混合仿真系統(tǒng)中由于包括了物理實(shí)驗(yàn)，收斂性問題是不能被接受的，因而采用顯式數(shù)值方法來有效避免隱式數(shù)值方法中存在的收斂性問題。此系統(tǒng)無(wú)須迭代，因此非常適合進(jìn)行物理實(shí)驗(yàn)。另外，此系統(tǒng)的開發(fā)無(wú)須改變有限元程序復(fù)雜且難以獲得的代碼，對(duì)于工程應(yīng)用很有吸引力。本系統(tǒng)采用了分離式實(shí)驗(yàn)系統(tǒng)框架。數(shù)值積分方法采用了Operator-Splitting（OS）方法。數(shù)據(jù)傳輸基于Socket網(wǎng)絡(luò)數(shù)據(jù)傳輸機(jī)制。 為了驗(yàn)證系統(tǒng)的有效性，本文利用該實(shí)驗(yàn)系統(tǒng)模擬了一個(gè)簡(jiǎn)單框架在地震下的響應(yīng)，并將此響應(yīng)與傳統(tǒng)的有限元分析方法進(jìn)行對(duì)比。對(duì)比結(jié)果證實(shí)了此系統(tǒng)的有效性和正確性。此外，本文還研究了影響實(shí)驗(yàn)系統(tǒng)有效性的重要參數(shù)，具體包括：時(shí)間步長(zhǎng)、加載時(shí)間、質(zhì)量調(diào)整以及材料阻尼。研究表明，為了獲得準(zhǔn)確的仿真結(jié)果，加載時(shí)間應(yīng)等于總時(shí)間且應(yīng)大于結(jié)構(gòu)周期2~3倍，質(zhì)量調(diào)整和材料阻尼有助于增大滿足穩(wěn)定條件的時(shí)間增量步，從而提高系統(tǒng)的計(jì)算速度。最后，研究中還運(yùn)用該系統(tǒng)模擬環(huán)梁節(jié)點(diǎn)的復(fù)雜鋼框架地震響應(yīng)，基于對(duì)試樣循環(huán)加載實(shí)驗(yàn)的結(jié)果，建立了環(huán)梁節(jié)點(diǎn)的精細(xì)有限元模型，并利用該有限元模型對(duì)環(huán)梁節(jié)點(diǎn)進(jìn)行了參數(shù)分析，使其能夠滿足強(qiáng)節(jié)點(diǎn)弱構(gòu)件的要求。模擬結(jié)果表明，本文提出的系統(tǒng)可以較好模擬采用了環(huán)梁節(jié)點(diǎn)的復(fù)雜鋼框架的地震響應(yīng)。
[Abstract]:Earthquake is one of the most destructive natural disasters. It is of great significance to accurately reproduce the seismic response of building structure to improve the seismic capacity of the structure and to prevent the building structure from collapsing in the earthquake. The conventional methods for reproducing structural seismic responses include numerical simulation and physical experiments. These two methods can usually obtain more accurate results, but both of them have some limitations. The hybrid on-line experimental technique can make full use of the advantages of numerical simulation and physical experiment, that is, the motion balance equation of structure can be established by using numerical simulation technique, and the restoring force characteristic of structure can be obtained by physical experiment. It is a promising technique to simulate the seismic response of structures accurately and effectively. In this paper, a new hybrid on-line experimental system combining finite element software and substructure technology is developed. This paper focuses on how to improve the simulation accuracy of numerical substructures in online hybrid experiments. The user subroutine interface (VDISP) is used as the interface between the main control program and the ABAQUS/Explicit module, and the target displacement is input to the ABAQUS/Explicit module through this structure and the reaction force is obtained. In this paper, the ABAQUS/Explicit module, not the ABAQUS/Standard module, is used because the complex structure is prone to convergence problems, and the convergence problem is unacceptable in the hybrid simulation system due to the physical experiments included. So the explicit numerical method is used to avoid the convergence problem in implicit numerical methods. The system does not need iteration, so it is very suitable for physical experiments. In addition, the development of the system does not need to change the finite element program complex and difficult to obtain code, it is very attractive for engineering applications. This system adopts the frame of separate experiment system. Operator-Splitting (OS) method is used in numerical integration method. Data transmission is based on Socket network data transmission mechanism. In order to verify the effectiveness of the system, this paper uses the experimental system to simulate the response of a simple frame under earthquake, and compares this response with the traditional finite element analysis method. The validity and correctness of the system are verified by the comparison results. In addition, important parameters affecting the effectiveness of the experimental system are studied, including: time step, loading time, mass adjustment and material damping. The results show that, in order to obtain accurate simulation results, the loading time should be equal to the total time and should be 3 times larger than the structural period. The quality adjustment and material damping can increase the time increment step to meet the stability conditions, thus improving the calculation speed of the system. Finally, the system is used to simulate the seismic response of the ring beam joints in complex steel frames. Based on the experimental results of cyclic loading on the specimens, the fine finite element model of the ring beam joints is established. The finite element model is used to analyze the parameters of ring beam joints, which can meet the requirements of strong joints and weak members. The simulation results show that the proposed system can better simulate the seismic response of the complex steel frame with ring beam joints.
【學(xué)位授予單位】：清華大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2013
【分類號(hào)】：TU317

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 汪訓(xùn)流;陸新征;葉列平;;往復(fù)荷載下鋼筋混凝土柱受力性能的數(shù)值模擬[J];工程力學(xué);2007年12期

本文編號(hào)：2155370