节点文献

海洋工程结构的参数识别及损伤诊断研究

Studies of Parameter Identification and Damage Detection for Ocean Engineering Structures

【作者】 王晓燕

【导师】 黄维平;

【作者基本信息】 中国海洋大学 , 港口、海岸及近海工程, 2004, 硕士

【摘要】 本文主要探讨了信息不完备情况下结构的参数识别及损伤诊断;论述了结构参数识别及损伤诊断的研究进展;讨论了遗传算法的发展现状、基本思想和特点;重点研究了海洋工程结构的参数识别和损伤诊断。 将缩减变量的卡尔曼滤波算法与最小二乘法相结合,用于结构的物理参数识别。在测量信息不完备的情况下,首先采用一种基于最小二乘法的统计平均方法得到系统的输入信息,然后将反演得到的结果作为输入,运用缩减变量的卡尔曼滤波算法对剪切型结构的物理参数进行识别及损伤诊断。数值算例表明,该算法具有较强的鲁棒性,识别精度较高。 针对遗传算法用于大型结构损伤诊断的弊端,本文将改进的遗传算法之一——CHC算法应用到损伤诊断领域,并在此基础上结合普通遗传算法的优点,提出了一种改进的CHC算法。以10单元固端梁和海洋平台为例,分别运用普通遗传算法和改进的CHC算法进行损伤诊断,对不同的损伤指标进行了详细的讨论;并针对不同准则下的传感器配置情况,对海洋平台的损伤诊断作了进一步探讨。数值算例表明,改进的CHC算法用于结构的损伤诊断能够减少计算时间,并具有收敛速度快,有一定的鲁棒性等特点。 在损伤诊断的结构分析过程中引入了逐阶滤频法。根据诊断的要求,通过逐阶滤频得到需要的损伤指标,而对其他的频率及振型则不予考虑。这一方法的引入优化了结构分析程序,减少运算时间,提高算法的效率。通过与普通分析程序对照,该方法的使用约能缩短90%的计算时间,解决了遗传算法的实际应用问题。

【Abstract】 This paper studies parameter identification and damage detection of structures in incomplete conditions, and discusses the development in parameter identification and damage detection of the structures, the basic theory, the advantages and the development of genetic algorithm. It focuses on parameter identification and damage detection of structures-specially, ocean offshore platforms.The method that combined REKF and least square method is used to detect the physical parameters of the structure. In the condition that detected information is partial, firstly system input which is used as the input of REKF method can be get by statistic average method that based on least square method, then REKF method is applied to identify physical parameters and diagnose the damage of shearing structures. The numerical example shows that, the method is of robustness, high precision and error tolerance.To solve the problem that it has a slow convergence when Genetic algorithm is used to diagnose the damage of large structures, CHC algorithm, an improved Genetic algorithm, is applied to the field of damage diagnosis in this paper, and an improved CHC algorithm (ICHC), which combines the advantages of simple genetic algorithm (SGA) and CHC algorithm, has been put forward. Both SGA algorithm and improved CHC algorithm are applied to diagnose the damage of a 10-unit fixed-end beam and an offshore platform, and different objective functions are discussed in details. This paper discusses the damage diagnosis of offshore platform with six different sensor placements, based on six different methods. Based on the numerical examples, it can be concluded that the calculating time in damage diagnosis of structures could be largely shortened by improved CHC method. Also it has the advantages of faster convergence and robustness.Step-frequency filter method is applied in the calculating of structural modal parameters. The application of this method optimizes the program of structural analysis, especially, it can shorten calculating time and has higher efficiency to the diagnosis of large structures. Comparing with the simple analysis program, it can save 90% time and solve the problem of the application of GA.

  • 【分类号】P75
  • 【被引频次】1
  • 【下载频次】217
节点文献中: 

本文链接的文献网络图示:

本文的引文网络