Surface Contamination Simulation and Control
Lead Research Organisation:
Loughborough University
Department Name: Aeronautical and Automotive Engineering
Abstract
This project will develop and test the capability to simulate surface water dynamics relevant to vehicle operation in adverse weather. Simple, relevant geometries will be used to generate experimental data and test new numerical methods. These will focus on new generic geometries representing, for example, a car wind screen, a-pillar and side glass with a focus on cases that capture the core physics.
Numerical simulation will explore the use of a new Coupled Level Set / Volume of Fluid (CLSVOF) approach allied to high-resolution aerodynamic models. Modelling of droplet contact properties will be explored - semi-empirical models of contact angle, effects of surface type, liquid properties (clean, variable solids fraction) along with contact models - splash, stick etc.
In addition, the work will explore the use of super-hydrophobic, super-hydrophilic and smart materials to control surface water deposition and movement without recourse to visually intrusive steps and channels.
Numerical simulation will explore the use of a new Coupled Level Set / Volume of Fluid (CLSVOF) approach allied to high-resolution aerodynamic models. Modelling of droplet contact properties will be explored - semi-empirical models of contact angle, effects of surface type, liquid properties (clean, variable solids fraction) along with contact models - splash, stick etc.
In addition, the work will explore the use of super-hydrophobic, super-hydrophilic and smart materials to control surface water deposition and movement without recourse to visually intrusive steps and channels.
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/N509279/1 | 01/10/2015 | 30/09/2020 | |||
| 1689977 | Studentship | EP/N509279/1 | 01/10/2015 | 30/09/2019 | Maciej Skarysz |
| Description | Introduction: Exterior water management (EWM) in the automotive industry involves predicting and controlling rainwater flow over vehicle surfaces. This is important for various reasons, such as driver safety, reliability of onboard systems based on optical sensors, braking performance and customer satisfaction. Car manufacturers are also obliged to study rainwater flow due to legal requirements. Research gap: In this project a CFD method is considered as a predictive tool for EWM, namely Coupled Level Set and Volume of Fluid (CLSVOF) for arbitrary meshes, which is a high fidelity, interface resolving method. The method is studied both as an alternative and complementary to the current state-of-the-art approach, i.e., the Lagrangian particle tracking and thin film modelling used by car manufacturers for EWM purposes. This approach, however, needs to make many assumptions which are often broken when realistic geometry is considered. Moreover, many physical phenomena such as droplet splash, breakup, or re-entrainment need to be modelled, often by empirical formulas, which are valid only for a certain range of parameters. Interface resolving methods, on the other hand, allow for highly accurate simulations of multiphase problems by a direct 3D representation of interface between fluids and solving physical equations for both phases, thus without need of additional modelling. This is obtained, however, at a high computational cost. Achievements: In this project, great attention is given to the computational cost and its reduction. An adaptive mesh refinement was implemented to CLSVOF method, and this was found to be the most critical factor for making such simulations feasible for EWM. A novel, iterative interface reconstruction method based on tetrahedral decomposition was proposed and implemented, and it was found to reduce overall simulation time by 1.3-1.6 compared to the existing widely used method. Further improvement was seen when a precursor simulation was used for single phase flow, which then was mapped into a smaller computational domain for multiphase simulation. A method to introduce resolved droplets into the multiphase simulation by boundary condition was proposed and a significant reduction in computational time was observed. The method, after proposed improvements, is feasible to be used in EWM. It can be applied for isolated cases such as droplet splash, droplet flowing into a drainage channel or rivulet flow. However, due to typical industrial simulation resources, it is not feasible to use it for multiple droplets impacts onto a large scale automotive geometry as a routine. Therefore, the method is likely to be complementary rather than a replacement for currently used methods. It was demonstrated on examples of droplet splashes and water re-entrainment that the method can be successfully used for derivation and validation of simplified models. Re-entertainment was studied in isolated conditions as well as on realistic geometry. The CLSVOF method was found to predict the Kelvin-Helmholtz instability accurately and this was validated against linear theory and experiments. The method was then used to validate a wave stripping model for a thin film method in realistic EWM conditions giving good agreement between CLSVOF results and the model. |
| Exploitation Route | The outcome of this project can be used by both academia and industry. For example, although the primary motivation for this project was EWM (automotive industry), the methods developed and presented here are equally applicable to many different areas, such as icing in the aerospace industry or fuel injection breakup process for IC engines or jet engines. Besides the usage of the method itself, there is a route related to the further development of the method by academia or industry producing engineering software. E.g., the adaptive mesh refinement was used with a fixed level of refinement for interfacial cell, whereas for regions of low curvature it might be sufficient to use coarser mesh. This might lead to problems with thin filaments though, which requires special attention. A significant improvement in computational time might be achieved by using local time-stepping, i.e., to update large cells less frequently than the refined cells. It will be especially important when high levels of refinement are used and differences in cell sizes are huge. Finally, a hybrid method is worth considering, e.g., transferring Eulerian secondary droplets generated during impact into a Lagrangian description. |
| Sectors | Aerospace, Defence and Marine,Chemicals,Energy,Manufacturing, including Industrial Biotechology,Transport |
| Description | In this project, a generic wing mirror geometry was studied by high fidelity method which led to derivation and validation of a simplified model which then was considered by industry. The research findings have been used by Jaguar Land Rover and Dassault Systèmes in the development of new models of PowerFLOW - engineering software for prediction of flow around vehicles. |
| First Year Of Impact | 2020 |
| Sector | Aerospace, Defence and Marine,Transport |