As computational methodologies become more integrated into industrial vehicle pre-development processes the potential for high transient vehicle thermal simulations is evident. This can also been seen in conjunction with the strong rise in computing power, which ultimately has supported many automotive manufactures in attempting non-steady simulation conditions.
The following investigation aims at exploring an efficient means of utilizing the new rise in computing resources by resolving high time-dependent boundary conditions through a series of averaging methodologies. Through understanding the sensitivities associated with dynamic component temperature changes, optimised boundary conditions can be implemented to dampen irrelevant input frequencies whilst maintaining thermally critical velocity gradients.
A sub-module derived from real vehicle geometry was utilised to evaluate a series of alternative averaging schemes (consisting of steady-state CFD points) in comparison to full CFD transient conditions. The size and simplicity of the model additionally allowed for an easy transition to the heavy computationally demanding unsteady conditions. The input data for both averaging schemes and full transient conditions were derived from the real vehicle driving profiles experimentally obtained on the Nuerburgring test track.
Qualitative analysis was conducted between the alternative schemes and full transient data in order to isolate the effects of dampening boundary conditions on consequent component temperatures. It was found that a weighted moving average can be optimal in resolving the high frequency changes whilst maintaining the average energy balance across under body components. The reactivity of the averaging schemes was dependent on the sampling rates in combination with the process of neutralising the inherent lag effects. Both these parameters had a significant effect on the time dependent results.
In order to isolate the effectiveness of the averaging schemes on the “warm-up phase”, multiple laps were conducted to locate the point at which temperature stabilisation occurs. Additionally the effect of incorrect initial component temperature was explored through evaluating the time taken to thermal stabilisation. It was found that under differing thermal conditions the time taken to thermal stabilisation was relatively constant regardless of the initialisation temperature.
The investigation explored the influence of improving the simulation accuracy by increasing the quantity of steady-state CFD points in locations of high velocity amplitudes over a short time period. Additionally the opposite was explored in regions of low velocity amplitude, in extended time phase. A relationship was found between the velocity gradient and the quantity of steady-state CFD points for component temperature. Through the findings of the investigation a Fourier type algorithm was exploited to further improve turnover efficiencies.