Research And Development Of Uncertainty Modeling And Analysis Module Based On Multi Discipline Optimization Platform

With the improvement of design level and the expansion of the application requirements, when people in the pursuit of high performance products at the same time the robustness and reliability of the product are also put forward higher and higher requirements. But in the process of product development, the process lack of knowledge, design error, manufacture error, the running environment of product and so on will produce many kinds of uncertainty. Products in the working process of the part of the performance indicators may produce change and migration, even serious deviation caused by the fault and failure may occur. In this paper, the modeling and analysis methods for uncertainty system research in a multidisciplinary process integration and optimization platform realizes the uncertainty analysis components research and development.Firstly, the fundamental theory of uncertainty multidisciplinary design optimization is studied. The paper involves the uncertainty of the basic conception of multidisciplinary design optimization. The sources and propagation of the uncertainty and its spread were summarized. The uncertainty can be divided into two types according to its sources and transmission: aleatory uncertainty and epistemic uncertainty.Secondly, the methods of uncertainty modeling technology is studied. Uncertainty multidisciplinary design optimization modeling including probability theory, convex models and interval theory, etc. On the basis, the uncertainty modeling module are analyzed and designed on Multidisciplinary platform in this paper. The uncertainty modeling technology focusing on the implements of the modeling method based on probability theory of uncertainty.Thirdly, the uncertainty analysis module is studied and implemented. The uncertainty analysis is an important part in uncertainty multidisciplinary design optimization. The uncertainty analysis is mainly studied how to according to the system input, the external environment and the uncertainty of the system itself influence to quantify the uncertainty of the system output. The uncertainty analysis focused on the multidisciplinary platform based on uncertainty analysis of Monte Carlo component implementation.Finally, this paper implements the uncertainty modeling and analysis module of research and development on the multidisciplinary design optimization platform FlowComputer, in addition, the relevant application examples are given in this paper.