Logo
Abstracts & Full Papers
303 - Dynamic reliability analysis of a stochastic structure with response surface method
Chen Y.
Abstract
Response Surface Method(RSM) for structural reliability analysis is simple, efficient and practical, which has been widely paid attention to by researchers. However, it was mainly used for static reliability analysis before. In this paper, RSM is introduced in the field of dynamic reliability analysis, based on which the method for dynamic reliability analysis of a stochastic structure is developed. Firstly, the root mean square of the dynamic response of a stochastic structure is considered as a variable changing with the structure parameters. Based on First Excursion Failure Criterion of random vibration and the performance function mode of random variables, the dynamic reliability performance function of the stochastic structure is established. Then RSM is used to fit the performance function. The quadratic function without cross terms, which is widely accepted and commonly used for its high fitting precision and calculation efficiency in most cases, is used as the fitting function. When calculating dynamic response with sampling points, such finite element analysis software as ANSYSˇ˘NASTRAN can be used to greatly simplify the calculation. Finally the reliability index is obtained through JC method. Calculation precision can be improved through sequential reconstitution of the fitting function. In the end of this paper, an example is presented to illustrate this method, which shows the method has the advantage of simplicity, generality, high precision and efficiency. So it has wide prospect of application especially in the dynamic reliability analysis of complex engineering structures.
Citation
Chen Y.: Dynamic reliability analysis of a stochastic structure with response surface method , CD-ROM Proceedings of the Thirtheenth International Congress on Sound and Vibration (ICSV13), July 2-6, 2006, Vienna, Austria, Eds.: Eberhardsteiner, J.; Mang, H.A.; Waubke, H., Publisher: Vienna University of Technology, Austria, ISBN: 3-9501554-5-7