基于分形理论关联维数的钢桁架损伤检测研究

发布时间：2018-05-03 11:59

本文选题：分形理论 + 关联维数　；参考：《河北科技大学》2014年硕士论文

【摘要】：近些年,钢结构被广泛用于大型工业、民业以及公共建筑设施中,但随着结构使用时间的延长,超负荷使用等问题的出现,使其出现损伤、失效等相关危害越来越多。因此,钢结构损伤检测不容忽视,寻找高效准确的检测方法也更受到研究人员的关注。针对钢结构独特结构形式,本文进行了以下几方面研究:1)本文介绍了分形理论、关联维数在国内外近些年的发展状况,对现阶段的理论研究进行了分类比较;引入了相空间重构、嵌入维数、时间延迟、无标度区等重要参变量概念,以一钢桁架为例,提出了基于分形理论关联维数的钢桁架结构的损伤检测方法。2)利用ANSYS软件建模,分别建立完好工况、单损伤工况、多损伤工况的桁架模型,并对结构进行瞬态分析,采集各工况下结构节点加速度响应信号。运用分形理论对结构损伤前后各节点响应信号直接进行关联维数处理,拟合了各节点ln(r)-(Cr)图形,并由此得出嵌入维数m与关联维数D的关系,最后对不同工况下节点关联维数差值绝对值△D进行分析比较,利用各工况节点△D的大小,达到确定杆件损伤位置和初步判别损伤程度的目的。3)考虑到实际检测过程中外界环境的影响,本文以环境噪声为例,模拟了上述桁架模型在实际环境中应用该方法进行损伤检测的情况。以随机函数形式,对桁架节点加速度响应信号施加15%随机水平噪声,对与噪声信号叠加后的节点振动信号重新进行关联维数处理,得出各节点新的关联维数△D值大小,与上述无噪声情况进行比较分析,确定该方法在噪声环境下进行钢桁架结构损伤检测时的抗噪声能力及稳定性。
[Abstract]:In recent years, steel structure has been widely used in large-scale industry, civilian industry and public building facilities. However, with the extension of service time of the structure and the emergence of overload problems, the damage and failure of steel structure are becoming more and more. Therefore, the damage detection of steel structures can not be ignored, and the search for efficient and accurate detection methods has been paid more attention by researchers. Aiming at the unique structure form of steel structure, this paper studies the following aspects: 1) this paper introduces the fractal theory, the development of correlation dimension at home and abroad in recent years, and compares the current theoretical research. The concepts of phase space reconstruction, embedding dimension, time delay and scale-free zone are introduced. Taking a steel truss as an example, the damage detection method of steel truss structure based on correlation dimension of fractal theory is proposed. The truss models of intact condition, single damage condition and multiple damage condition are established, and the transient analysis of the structure is carried out, and the acceleration response signals of the structure node are collected under each working condition. Using fractal theory, the correlation dimension of the response signals of each node before and after structural damage is directly processed, and the relationship between the embedded dimension m and the correlation dimension D is obtained. Finally, the absolute value D of the node correlation dimension difference under different working conditions is analyzed and compared. In order to determine the damage location and judge the damage degree of the member. 3) considering the influence of the outside environment in the actual detection process, this paper takes the environmental noise as an example. The application of the method to the damage detection of the truss model is simulated. In the form of random function, 15% random horizontal noise is applied to the acceleration response signal of the truss node, and the correlation dimension of the joint vibration signal superimposed with the noise signal is processed again, and the new correlation dimension D value of each node is obtained. Compared with the above noise-free condition, the anti-noise ability and stability of the method are determined when the damage detection of steel truss structures is carried out under the noise environment.
【学位授予单位】：河北科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TU317;TU391

【参考文献】