Developing a methodology for the evaluation of results uncertainties in CFD codes

In this work the development of a methodology is studied to evaluate the uncertainty in the results of CFD codes and is compatible with the VV-20 standard Standard for Verification and Validation in CFD and Heat Transfer ^, developed by the Association of Mechanical Engineers ASME . Similarly, the alternatives are studied for obtaining existing uncertainty in the results to see which is the best choice from the point of view of implementation and time. We have developed two methods for calculating uncertainty of the results of a CFD code, the first method based on the use of techniques of Monte-Carlo for the propagation of uncertainty in this first method we think it is preferable to use the statistics of the order to determine the number of cases to execute the code, because this way we can always determine the confidence interval desired level of output quantities. The second type of method we have developed is based on non-intrusive polynomial chaos. (Author)