摘要
以某12层钢筋混凝土标准框架结构振动台模型试验的数据为基础,用卡尔曼滤波技术与移动谱相结合的方法对该标准框架结构参数的时变特性进行研究,该方法相当于对时域信号施加变长度矩形窗后进行卡尔曼滤波,有效地提高了参数识别精度。模型结构在小震激励时,结构的频率发生微小程度的降低,表明结构内部发生了未表现出来的轻微损伤;大震激励时,结构频率较初始时刻发生大幅度的下降,表明结构发生了较大程度的破坏。经与理论解和分量分析识别结果对比,用本文方法识别出的结构参数更接近理论解。
Based on the shaking table model test data of a 12-story RC-frame structure, the time-vatTing characteristics of standard frame structural parameters are researched in this paper by using the combination of Kalman filter and mobile spectrum. The results are precise. It is found that the frequency gets a very slight reduction when the model structure is under the excitation of small earthquakes. This indicaties that tiny damage which is notfounded is located in the structure. The frequency of the structure is greatly reduced compared to that at the initial moment under the large earthquake excitation. It is shown that there is severe damage in the structure. By the comparison with component analysis and theoretical solution, the recognition results by using the proposed method is closer to the theoretical solution.
出处
《地震工程与工程振动》
CSCD
北大核心
2013年第1期40-46,共7页
Earthquake Engineering and Engineering Dynamics
基金
国家自然科学基金项目(50908027)
地震工程与工程振动开放实验室项目(2009A10)
关键词
卡尔曼滤波
移动谱
参数时变特性
模型试验
Kalman filter
mobile spectrum
time-varying characteristics
model test
