Application of the DSMC method to high density micro-flows.

摘要

This research discusses the application of the Direct Simulation Monte Carlo (DSMC) method to flows in Micro Electro Mechanical devices (MEMS) and the Space Shuttle leading edge reinforced Carbon-Carbon destruction due to the chemically active flows. For the MEMS device application an analysis to optimize the shape and the flow parameters of the micronozzle has been provided in order to increase the device performance. For the Space Shuttle wing problem a technique which predicts the microchannel-crack wall destruction rate has been proposed. The technique helps to ensure the safety of the Space Shuttle orbiter and future complex reentry platforms. Careful study of these complex flows requires a comprehensive suite of tools composed of flow solvers capable of accurately predicting the properties of high gradient flows for a wide range of the flow regimes. In this research a robust Direct Simulation Monte Carlo method (DSMC) in combination with its equilibrium version, designated as eDSMC, has been applied to model these two types of internal flows.; A second aspect of the research has to do with an analysis of the errors that are present in the results of a statistical method. In the case of DSMC two types of errors have been distinguished and separately analyzed: statistical errors which naturally occur in a DSMC solution and deterministic errors which depend on the correct choice of the numerical parameters of the scheme, such as time step and cell sizes.; Before considering the application to a real case the proposed computational techniques have been extensively evaluated through a number of test cases, which can be compared with data or previously available numerical solutions. In particular a micronozzle flow has been solved and the results have been compared with the experimental data from Ref. [1] and a microchannel flow for which a solution is also available in Ref. [2]. When possible and practical the obtained solutions are compared with the usual Naiver Stokes results. Different computational techniques used in this research complement each other and provide better inside into the considered problems. The results of the application of the proposed techniques to the real problems shows the potential of the statistical methods when they are applied to the multiscale flow problems. In this research two complex problems have been solved: optimization of micronozzle performance and prediction of micro channel expansion for bare carbon materials exposed to high energy reentry flows. In both cases the numerical results show good agreement with the experimental data obtained at NASA laboratories.; The presented research demonstrates the ability of the proposed statistical methods to model complex 2D and 3D flow configurations which include complex, chemically active boundary conditions. The proposed statistical schemes are inherently stable, computationally effective, and able to employ adaptive computational mesh structure. In many cases they provide further benefits over traditional CFD schemes by simplifying the problem setup.; Further developments of the presented techniques are suggested. In particular a combination of the baseline DSMC and the proposed eDSMC scheme suggests a potential method for effectively solving flows that are multiscale in terms of the Knudsen and Reynolds numbers and exhibit complex physical and chemical phenomena.