Aerodynamic Impact of Composite Wheel Covers on Electric Vehicles-2

Part 2

1. Aerodynamic Structure

The application of renewable energy sources to the transportation sector and the increased use of electric vehicles have introduced problems such as storage efficiency and range. For this reason, the development of electric vehicles with more aerodynamic and lighter designs has become a priority in the design process. This section explains the numerical approach used to simulate tire rotation. Computational fluid dynamics (CFD) codes offer different numerical approaches to define models with rotating parts. These are: Rotating Wall (RW), Moving Reference Frame (MRF), and Sliding Mesh (SM). These approaches are explained in two ways in this section to introduce the MRFg (Moving Reference Frame - Grooves) approach. The analysis of MRFg performance is conducted in two steps. First, it is compared with sliding mesh simulations on an isolated wheel, and subsequently, the aerodynamic effects on the drag coefficient are analyzed in detail on the complete vehicle.

1.1. Model, Domain and Boundary Conditions

Figure 4. Wheel design of Demobil 09 electric vehicle

Figure 5. Wheel cover designs

Prior to the study, wheel covers were designed as shown in Figure 5 with the aim of contributing not only to the aerodynamic structure but also to the visual appearance of the wheel design, and their use was deemed appropriate.

Figure 6. Test field

In the first phase of CFD analysis preparation, the dimensions of the boundary volume and boundary conditions are determined. The volume was set at 15 [m] length, 4 [m] width, and 3 [m] height. Since Demobil 09 is a symmetric vehicle, the volume was designed symmetrically from the center to shorten analysis time and obtain results closer to experimental studies.

Table 4. Boundary conditions

1.2. Solution Mesh Generation (Meshing)

The CFD problem in question consists of bodies in static (vehicle body) and dynamic (wheel) conditions. For this reason, the problem was solved with transient behavior in terms of time, and a dynamic mesh structure was applied as the solution mesh structure. To overcome the challenge of creating a healthy solution mesh structure, the most realistic modeling method is to move the mesh completely. For this reason, a transient and physically rotating mesh structure at each time step is required. This is generally known as the sliding mesh (SM) method and is applied as rigid body motion. For this reason, it is easily applicable to wheels. However, the mesh moving at each time step and connecting it to neighboring fixed cells presents us with high computational time and leads to a significant increase in working time. It was determined that the analysis convergence time, solution mesh transformation time, and parallelization of the number of cells depend greatly on the magnitude. In this study, it was planned to examine the change in drag coefficient of the electric vehicle as a result of using wheel covers designed for wheels used on an electric vehicle. During the analyses, the speed of the electric vehicle was set at 120 [km/h] and the wheel rotational speed was set at 1140 [RPM].

Table 5. Mesh structure settings

In this analysis, the wheel rotates and therefore can be modeled using a sliding mesh (SM) approach. Since SM physically shifts the mesh structure at each time step, it performs the simulation of actual wheel rotation and is considered a correct rotation modeling method for CFD applications. In this study, a single tetrahedral mesh structure was created for each wheel such that all rotation modeling procedures could be implemented, thus ensuring that there would be no repeatability resulting from the solution mesh structure. The specified solution mesh settings were created to obtain the closest results while keeping mesh structure size and computational costs low. The largest cell size is 256 [mm] at the field boundaries, and the general mesh size was approximately 3.2 million cells.

1.3. Aerodynamic Drag Force Acting on the Vehicle

As a result of the analyses conducted, the calculation of the aerodynamic drag force acting on a moving vehicle is specified in equation (1). In the equation, ρ represents the air density, A the vehicle's frontal area, V the velocity, and CD the vehicle's drag coefficient. When the equation is examined, it is observed that the aerodynamic drag force acting on the vehicle increases with the square of velocity. The calculation of the power spent by the vehicle to overcome the aerodynamic drag force is specified in equation (2).

Table 6. Interpolation settings

Table 7. Criteria

2. Results and Discussion

Table 8. Analysis results

The results obtained from isolated wheel and full vehicle studies are presented and discussed in this section. As a result of the analyses conducted, theoretical studies determined that when wheel covers are used on Demobil 09 traveling at a speed of 120 [km/h], energy consumption would be reduced by 430 [W] per hour.

Figure 8. Pressure distribution on the body, vortex core region, and velocity vector analysis as a result of Demobil 09 EV analysis without wheel covers

Figure 9. Pressure variation along the green line

In the CFD analyses of Demobil 09 conducted without wheel covers, the images found in Figures 8, 9, and 10 were examined. In the pressure variation graph on the line indicated in Figure 9, a significant pressure difference occurred in the wheel area of the vehicle when there was no wheel cover. When there was a wheel cover, this pressure difference decreased by approximately 100 [Pa]. The effects of the pressure drop are clearly visible in the vortex formation regions and velocity vectors indicated in Figure 10.

Figure 10. Pressure distribution on the body, vortex formation regions, and velocity vector analysis as a result of the analysis conducted with Demobil 09 EV wheel covers

Figure 11. Wheel cover and its installed state on the wheel

As a result of the studies conducted, the product obtained from the female molds of the design specified in Figure 2 and its installed state on the related wheel is shown in Figure 11.

3. Acknowledgments

The authors wish to thank Cevher Jant Sanayi A.Ş., colleagues from the Solaris Solar Car Team, and Burak Kurttay for their assistance in the design, manufacturing, and performance testing. Dr. Aytaç Gören Dokuz Eylül Üniversitesi Department of Mechanical Engineering Automatic Control and Robotics Laboratories   Talha Batuhan Korkut Dokuz Eylül Üniversitesi Department of Mechanical Engineering Automatic Control and Robotics Laboratories   Onur Özaydın R&D Engineer Cevher Jant Sanayii A.Ş.     Elvan Armakan R&D Engineer Cevher Jant Sanayii A.Ş.    

4. References [1] Talha Batuhan Korkut, Aytaç Gören, Mehmet Akif Ezan. June 2018. "Photovoltaic Panel Efficiency Modelling for Solar Powered Vehicle", 7th Global Conference on Global Warming (GCGW-2018), June 2018, Izmir (Turkey) [2] Talha Batuhan Korkut, Ahmet Ozkan, Aytaç Gören. March 2018."Use of Composite Sheets in Electric Vehicles",PUTech&Composites 8 (38), 8-14 [3] Aytaç Gören, Özgür Başer. November 2007. "Body Design and Power Requirement in Solar Powered Vehicles", Makinatek, Issue 121, Pages 124-129 [4] Aytaç Gören, Yusuf Can Arslan. April 2015, "Use of Mold Material in Polymer Composite Vehicle Body Production", PUTech&Composites [5] Christoffer Landstrom, Linda Josefsson, Tim Walker, Lennart Löfdahl. April 2012, "Aerodynamic Effects of Different Tire Models on a Sedan Type Passenger Car", SAE International, doi: 10.4271/2012-01-0169 [6] Christoffer Landstrom, Tim Walker, Lasse Christoffersen, Lennart Löfdahl. December 2011, "Influences of Different Front and Rear Wheel Designs on Aerodynamic Drag of a Sedan Type Passenger Car", SAE International, doi: 10.4271/2011-01-0165 [7] Andrew D'Hooge, Robert B.Palin, S.Johnson, Bradley Duncan, Joaquin Ivan Gargoloff. April 2012, "The Aerodynamic Development of the Tesla Model S – Part 2: Wheel Design Optimization", SAE International, doi: 10.4271/2012-01-0178 [8] Alexander Waschle. 2007, "The Influence of Rotating Wheels on Vehicle Aerodynamics – Numerical and Experimental Investigations", SAE Technical Paper Series, ISBN 0-7680-1631-2 [9] Aytaç Gören, Cesim Ataş. December 2008, "Manufacturing of Polymer Matrix Composites Using Vacuum Assisted Resin Infusion Molding", Archives of Material Science and Engineering, Volume 34, Issue 2, Pages 117-120