参数信息不完备状况下桥梁结构损伤识别方法研究

发布时间：2018-04-17 13:47

本文选题：桥梁结构 + 损伤识别　；参考：《吉林大学》2015年硕士论文

【摘要】：桥梁作为重要的基础设施组成部分,在国家经济、社会发展中起到了不可替代的作用。随着综合国力的不断加强,桥梁建设更是取得了举世瞩目的成就。但是,随着车辆荷载的不断增加与材料劣化的日益加剧,桥梁损伤及失效不可避免。对现役桥梁及旧桥进行健康监测及损伤识别,对保证桥梁的安全运营具有重要的实际意义。基于振动的损伤识别方法是通过动力参数的改变来辨识和识别损伤。然而,损伤识别方法在实际应用中受到测试噪声、测试信息不完备、基准模型缺失等不确定因素的影响,造成损伤识别参数获取不完整,降低了损伤识别结果的可靠性。如何解决参数信息不完备状况下的损伤识别,成为促进桥梁损伤识别方法实际应用的关键。因此,本文针对测试参数误差、信息缺失及基准模型缺失等不确定性状况,基于模糊理论与数据拟合重构技术,提出了具有较强鲁棒性的损伤识别方法。具体开展了以下工作:(1)详细介绍参数信息不完备状况下的桥梁损伤识别研究背景及意义;阐述了传统损伤识别方法在处理不确定问题中的不足。总结了国内外损伤识别研究现状,提出了本文的主要研究内容。(2)阐述了粒子群优化理论与模糊C均值聚类算法的基本原理,提出了粒子群优化模糊C均值聚类算法。考虑到桥梁结构损伤识别指标在相近损伤状态中的相似性,将粒子群优化模糊C均值聚类方法用于结构损伤识别,建立了基于粒子群优化模糊C均值聚类的桥梁损伤识别方法和流程。(3)针对简支梁测试参数误差及部分数据缺失的信息不完备状况,提出了基于粒子群优化模糊C均值聚类算法的损伤识别方法。在原始数据中添加高斯白噪声,以振型数据缺失模拟信息不完备,采用PSO-FCM方法进行损伤识别。损伤识别结果验证了该方法处理参数误差及数据缺失状况损伤识别的有效性和准确性。(4)采用多项式拟合对基准模型均匀荷载面曲率参数进行重构,得到未损伤结构的均匀荷载面曲率值,利用均匀荷载面曲率差进行损伤定位。在损伤定位的基础上,采用PSO-FCM方法确定损伤程度。有限元仿真验证了方法在基准模型缺失下的损伤定位与损伤程度识别的准确性。
[Abstract]:As an important infrastructure component, bridge plays an irreplaceable role in national economic and social development.With the continuous strengthening of comprehensive national strength, bridge construction has made remarkable achievements.However, with the increasing of vehicle load and deterioration of materials, bridge damage and failure are inevitable.Health monitoring and damage identification of existing bridges and old bridges are of great practical significance to ensure the safe operation of bridges.The damage identification method based on vibration is to identify and identify the damage by changing the dynamic parameters.However, the damage identification method is affected by some uncertain factors, such as test noise, incomplete test information and lack of reference model, which result in incomplete acquisition of damage identification parameters and decrease the reliability of damage identification results.How to solve the problem of damage identification under the condition of incomplete parameter information has become the key to promote the practical application of bridge damage identification method.Therefore, based on fuzzy theory and data fitting reconstruction technology, a robust damage identification method is proposed in this paper, which is based on the uncertainty of test parameter error, information loss and the absence of reference model, and is based on fuzzy theory and data fitting reconstruction technology.The research background and significance of bridge damage identification under the condition of incomplete parameter information are introduced in detail, and the shortcomings of traditional damage identification methods in dealing with uncertain problems are expounded.In this paper, the present situation of damage identification at home and abroad is summarized, and the main research content of this paper is put forward. The theory of particle swarm optimization and the basic principle of fuzzy C-means clustering algorithm are expounded, and the fuzzy C-means clustering algorithm of particle swarm optimization is put forward.Considering the similarity of damage identification indexes of bridge structures in similar damage states, particle swarm optimization (PSO) fuzzy C-means clustering method is applied to structural damage identification.A method and flow chart for bridge damage identification based on particle swarm optimization (PSO) fuzzy C-means clustering is established.A damage identification method based on particle swarm optimization fuzzy C-means clustering algorithm is proposed.Gao Si white noise is added to the original data, and the simulation information of mode data is missing. The PSO-FCM method is used to identify the damage.The result of damage identification verifies the validity and accuracy of the method in dealing with parameter errors and data loss. The polynomial fitting is used to reconstruct the curvature parameters of the uniform load surface of the benchmark model.The curvature value of uniform load surface of undamaged structure is obtained, and the damage location is carried out by using the curvature difference of uniform load surface.On the basis of damage location, PSO-FCM method is used to determine the degree of damage.Finite element simulation verifies the accuracy of damage location and damage degree identification in the absence of reference model.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U446

【参考文献】