A new methodology for the CFD uncertainty analysis

Abstract

With respect to the measurement uncertainty, this paper discusses the definition, the sources, the classification and the expressions of the CFD uncertainty. Based on the orthogonal design and the statistics inference theory, a new verification and validation method and the related procedures in the CFD simulation are developed. With the method, two examples of the CFD verification and validation are studied for the drag coefficient and the nominal wake fraction, and the calculation factors and their interactions which would significantly affect the simulation results are obtained. Moreover, the sizes of all uncertainty components resulting from the controlled and un-controlled calculation factors are determined, and the optimal combination of the calculation factors is obtained by an effect estimation in the orthogonal experiment design. It is shown that the new method can be used for the verification in the CFD uncertainty analysis, and can reasonably and definitely judge the credibility of the simulative result. As for CFD simulation of the drag coefficient and the nominal wake fraction, the results predicted can be validated. Although there is still some difference between the simulation results and the experiment results, its approximate level and credibility can be accepted.