Stochastic particle tracking models utilising RANS simulations are frequently used to simulate particulate matter damage in gas turbines. The Continuous Random Walk (CRW) model has been shown to model particulate motion in the diffusion-impaction regime significantly more accurately than Discrete Random Walk implementations in simple flows, however, the model uses turbulent flow statistics determined from DNS of flow through channels and experiments in pipes. No published work is available to allow these CRW models to be used in highly accelerating flows. Validation of the models in accelerating flow is important to enable the model to be used in design studies of rotating engine-realistic geometries of complex curvature.

Previous work by the authors has shown that using RSM predicted statistics is a promising approach, with more accurate predictions being made whilst being less reliant on DNS extracted statistics. Reynolds stress distributions and time scales are extracted from the fluid solution, the former are corrected based on known discrepancies, when compared to DNS, in simpler flows. Changes are also suggested to the calculation of the time step. The effect of these changes were studied using a commercial CFD solver using a high density mesh of a 90 degree bend, for which significant literature exists and the importance of capturing turbulent dispersion has been reported. Here further improvements are made to account for Reynold Stresses cross-terms, remove any requirement for the time step to be limited by a maximum or minimum value, and remove any dependence of DNS extracted statistics.

This new model is applied and validated in a channel flow, a pipe flow, and a 90 degree bend. This is essential in showing that the presented CRW model is properly implemented and has good predictive capability across a range of flows. By starting in simpler flows the implementation can be verified more easily due to the existence of flow symmetry planes. Due to implementation using User-defined Functions (UDFs) and use of a Body-fitted co-ordinate system there are challenges in calculating gradients of Reynolds stresses from the fluid solution in Fluent. Here, the calculation of these gradients is validated and compared against either openly available or simulated DNS (Channel Flow, Bend at Re = 1000 and 6000) and LES (Pipe and Bend at Re=10000). Particle tracking results of this model are validated against experimental and computational results in pipe and bend flows at multiple Reynolds numbers using bulk non-dimensional deposition velocities for a range of non-dimensional particle relaxation times. In the case of the bend spatially resolved information is also available as a computational result. The ability to estimate the convergence of spatially resolved stochastic particle tracking simulations is shown, defined, and used. The presented extension of the CRW capability is a step towards enabling it to be used on more realistic geometries and flow conditions.


Development and Validation of a Continuous Random Walk Model for Particle Tracking in Accelerating Flows

Division / Sponsor:
International Gas Turbine Institute (IGTI)

Paper Proceedings: https://asmedigitalcollection.asme.org/GT/GT2020/volume/84058

Turbo Expo - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.