含裂缝的简支梁桥动力特性分析及参数识别研究

发布时间：2018-04-29 10:50

本文选题：简支梁 + 动力特性　；参考：《吉林大学》2017年硕士论文

【摘要】：混凝土桥梁的损伤通过裂缝表现出来,在混凝土桥梁服役期内,裂缝的出现无法避免。因此,分析了解开裂梁的动力特性并进行损伤识别意义重大。目前,基于振动的损伤识别已被推广成为桥梁健康监测系统中极其重要的一部分。进行开裂梁分析时,通常有两种裂缝模拟方式:张开裂缝模型以及呼吸裂缝模型。实际状态中,由于梁体荷载作用等原因,裂缝在振动过程中发生交替开合现象,呼吸裂缝模型更为合理。本文开展的主要研究工作包括:首先,从张开裂缝模入手,采用分离弹簧法进行开口裂缝模拟,分析开口裂缝的存在对简支梁动力特性的影响。分析结果表明具有裂缝的简支梁与完整梁的动力特性有所不同,完整梁在振动过程中各阶固有频率均为定值,而裂缝的出现使简支梁振动频率减小,裂缝深度与裂缝所在位置均能影响开裂梁的动力特性。基于神经网络理论,以频率参数构造损伤识别指标,实现了开口裂缝梁的损伤位置和程度识别。此外,采用有限元仿真方式模拟开口裂缝,分析了模态曲率、模态曲率差、模态柔度差、模态柔度差曲率四个损伤识别指标的开口裂缝识别效果。基于神经网络拟合方法,实现了不依赖基准模型数据的简支梁损伤识别。然后,选用呼吸裂缝模型,对现行的呼吸裂缝时频分析方法进行了分析和比较,选取Hilbert-Huang变换对开裂简支梁进行时频分析,得到简支梁的Hilbert谱等动力响应。分析结果表明具有呼吸裂缝的简支梁的自振频谱在基频的整数倍位置上出现了高次谐波成分,呼吸裂缝的出现令梁的振动呈现出明显的非线性特征。简支梁的基频随着裂缝相对深度的增加而减小,而高次谐波的阶数与幅值则随损伤程度的加剧而增加。简支梁开裂之后其瞬时频率不再是常数,而是在某一范围内上下浮动,且变化范围随着损伤程度的加剧而增大。最后,通过EMD提取出瞬时频率并对数据进行统计分析,提出利用瞬时频率相对差值、瞬时频率方差、正态峰值比以及频谐比等损伤识别指标的概念,并证明了各指标与α(裂缝相对深度)之间遵循多项式的关系,均具有初步识别呼吸裂缝的能力。同时也发现频谐比对裂缝相对位置具有较高的灵敏程度,能够初步识别裂缝位置。
[Abstract]:The damage of concrete bridges is manifested by cracks, and the cracks can not be avoided during the service period of concrete bridges. Therefore, it is of great significance to analyze the dynamic characteristics of cracked beams and identify the damage. At present, Vibration-based damage identification has been widely used as an extremely important part of bridge health monitoring system. In the analysis of cracked beams, there are usually two kinds of crack simulation methods: open crack model and respiratory crack model. In the actual state, because of the action of beam load, the phenomenon of alternate opening and closing occurs in the vibration process of the crack, and the model of respiratory crack is more reasonable. The main research work in this paper is as follows: firstly, starting with the opening crack model, the split spring method is used to simulate the opening crack, and the influence of the existence of the opening crack on the dynamic characteristics of the simply supported beam is analyzed. The analysis results show that the dynamic characteristics of simply supported beams with cracks are different from those of intact beams. The natural frequencies of each order of intact beams are fixed in the process of vibration, and the vibration frequency of simply supported beams is reduced by the occurrence of cracks. The depth of crack and the location of crack can affect the dynamic characteristics of cracked beam. Based on the theory of neural network, the damage identification index is constructed with frequency parameters, and the damage location and degree of open crack beam are identified. In addition, the open crack is simulated by finite element simulation, and the open crack identification effect of four damage identification indexes, namely modal curvature, modal curvature difference, modal flexibility difference and modal flexibility difference curvature, is analyzed. Based on the neural network fitting method, the damage identification of simply supported beam is realized, which does not depend on the datum model data. Then, the existing time-frequency analysis method of respiratory crack is analyzed and compared with the respiratory crack model. The time-frequency analysis of simply supported beam with Hilbert-Huang transformation is carried out, and the dynamic response of Hilbert spectrum of simply supported beam is obtained. The analysis results show that the frequency spectrum of the simply supported beam with respiratory crack has high harmonic component at the integer position of the fundamental frequency, and the vibration of the beam is obviously nonlinear due to the occurrence of the respiratory crack. The fundamental frequency of simply supported beam decreases with the increase of the relative depth of crack, while the order and amplitude of higher harmonic increase with the increase of damage degree. The instantaneous frequency of the simply supported beam after cracking is no longer constant, but fluctuates up and down in a certain range, and the range of variation increases with the severity of damage. Finally, the instantaneous frequency is extracted by EMD and the data are statistically analyzed. The concept of damage identification index, such as instantaneous frequency relative difference, instantaneous frequency variance, normal peak value ratio and frequency harmonic ratio, is put forward. It is proved that the relation between each index and 伪 (relative depth of fracture) follows polynomial, and all of them have the ability to identify the respiratory fissure preliminarily. At the same time, it is also found that the frequency harmonic ratio has a high sensitivity to the relative position of the fracture, and it can be used to identify the fracture position.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U446

【相似文献】