Search Results

A simulation framework for vehicle aerodynamics using up to 10 billion fully unstructured cells has been developed on a world-fastest class supercomputer, called the K computer, in Kobe, Japan. The simulation software FrontFlow/red-Aero was fully optimized on the K computer to utilize up to 10,000 processors with tens of thousands of cores. A hybrid parallelization method using MPI among processors and OpenMP among cores inside each processor was adopted. The code was specially tuned for unsteady aerodynamic simulation including large-eddy simulation, and low Mach number approximation was adopted to avoid excessive iterations usually required for the fully incompressible algorithm. The automated mesh refining system was developed to generate unstructured meshes of up to 10 billion cells. In the system, users only generate unstructured meshes in the order of tens of millions of cells directly using commercial preprocessing software.

The aerodynamic performance of new vehicles is commonly determined using computational fluid dynamics (CFD) and wind tunnel tests. The final assessment is carried out by actual running tests. In particular, ideas regarding fuel consumption improvement that relate to components for the reduction of the coefficient of drag (CD) value are evaluated by coast-down tests. However, a difference often exists between the component's efficiency between wind tunnel tests and coast-down tests. Therefore, we focused on the efficiency of an air-dam spoiler in reducing CD values. A comparison was made between the aerodynamic effect of the air-dam spoiler in wind tunnel and coast-down tests in terms of the CD value and the wake structure behind the vehicle. To determine the relationship between the CD value and the wake structure behind the vehicle, we measured vehicle speed, wind velocity and direction, vehicle height, and pressure distribution on the back door.

The critical change in drag occurring on the Ahmed body when the slanted base has an angle of 30° is due to a transition in the wake structure. In a previous study on flow analysis across the Ahmed body, we investigated the unsteady wake experimentally using hot-wire and particle image velocimetry measurements. However, because the experimental analysis yielded limited data, the spatially unsteady wake behaviour, interaction between the trailing vortex and transverse vortices (up/downwash), and flow mechanism near the body were not discussed sufficiently. In this study, the unsteady wake structures were analysed computationally using computational fluid dynamics to understand these issues, and the hypothesis was tested. The slant angle was 27.5°, which is identical to that in the experiment and corresponds to a high drag condition indicated experimentally.

One-way coupled simulation method that combines CFD, structural and acoustical analyses has been developed aiming at predicting the aeroacoustical interior noise for a wide range of frequency between 100 Hz and 4 kHz. Statistical Energy Analysis (SEA) has been widely used for evaluating transmission of sound through a car body and resulting interior sound field. Instead of SEA, we directly computed vibration and sound in order to investigate and understand propagation paths of vibration in a car body and sound fields. As the first step of this approach, we predicted the pressure fluctuations on the external surfaces of a car by computing the unsteady flow around the car. Secondly, the predicted pressure fluctuations were fed to the subsequent structural vibration analysis to predict vibration accelerations on the internal surfaces of the car.

A wall-resolving Large Eddy Simulation (LES) has been performed by using up to 40 billion grids with a minimum grid resolution of 0.1 mm for predicting the exterior hydrodynamic pressure fluctuations in the turbulent boundary layers of a test car with simplified geometry. At several sampling points on the car surface, which included a point on the side window, the door panel, and the front fender panel, the computed hydrodynamic pressure fluctuations were compared with those measured by microphones installed on the surface of the car in a wind tunnel, and effects of the grid resolution on the accuracy of the predicted frequency spectra were discussed. The power spectra of the pressure fluctuations computed with 5 billion grid LES agreed reasonably well with those measured in the wind tunnel up to around 2 kHz although they had some discrepancy with the measured ones in the low and middle frequencies.

In open-grille vehicle aerodynamics simulation using computational fluid dynamics, in addition to basic flow characteristics, such as turbulent flow with a Reynolds number of several million on the bluff body, it is important to accurately estimate the cooling air flow introduced from the front opening. It is therefore necessary to reproduce the detailed geometry of the entire vehicle including the engine bay as precisely as possible. However, there is a problem of generating a good-quality calculation grid with a small workload. It usually takes several days to a week for the pretreatment process to make the geometry data ‘clean’ or ‘watertight’. The authors proposed a computational method for complex geometries with a hierarchical Cartesian grid and a topology-independent immersed boundary method with dummy cells that discretize the geometry on a cell-by-cell basis and can set an imaginary point arbitrarily.

For better performance of outboard motors for high speed craft, improvement in the performance of the propeller strut located ahead of the propeller is indispensable in addition to ameliorating the performance of the screw propeller itself. Thus, it is extremely important to reduce the drag of the propeller strut, which accounts for the predominant portion of the submerged parts of the motor and hull when the craft is running at high speed and to improve the propeller efficiency in the wake of the propeller strut. This paper, taking up two different shapes of the propeller strut, compares the performances of the propeller placed in the wake of the propeller strut in tank tests, and discusses the drag of the propeller strut. The two propeller strut shapes are that of a 70% scaled down model of the propeller strut SUZUKI's 200 PS outboard motor and its improved version.

Reduction of flow resistance in an intake system is essential for increasing the output of a four-stroke engine. Evaluation method regardless engine displacement or number of valves or cylinder must be required in intake system design. This study proposes intake loss coefficient as total evaluation method from flow in an intake port to charging flow into a cylinder. A three-dimensional, general-purpose Computational Fluid Dynamics (CFD) code was used to calculate an intake loss coefficient. A correlation was confirmed between an intake loss coefficient and the engine power output. Intake loss coefficients and the CFD technique may be used for efficient optimization of the shape of an intake system.

When the planing craft with outboard motor is running, cavitation occurs around the surface of propeller and lower unit of outboard motor. Cavitation has been classified under several categories by the feature and cause of occurrence. Among them, cloud cavitation and root cavitation lead to erosion damage on the surface of lower unit and propeller. To prevent from poor appearance or performance deterioration of outboard motor by erosion damage, it is important problem to predict the erosion occurrence. Currently we can predict the cavitation phenomena sufficiently, but the area of cavitation does not necessarily correspond with the area of erosion. In this study, we present the new method to predict the area of erosion due to cavitation using CFD (computer fluid dynamics) analysis. In order to evaluate the accuracy of erosion occurrence simulation, the simulation results are compared against the result of a full-scale cruising test.

The objective of this paper is to investigate the potential of lean burn combustion to improve the thermal efficiency of spark ignition engine. Experiments used a single cylinder gasoline spark ignition engine fueled with primary reference fuel of octane number 90, running at 4000 revolution per minute and at wide open throttle. Experiments were conducted at constant fueling rate and in order to lean the mixture, more air is introduced by boosted pressure from stoichiometric mixture to lean limit while maintaining the high output engine torque as possible. Experimental results show that the highest thermal efficiency is obtained at excess air ratio of 1.3 combined with absolute boosted pressure of 117 kPa. Three dimensional computational fluid dynamic simulation with detailed chemical reactions was conducted and compared with results obtained from experiments as based points.