Abstract

Abstract In full vehicle thermal flow analyses, the most often used procedure to simulate fluid motions driven by the cooling fan is the Moving Reference Frame (MRF) method. In the MRF approach, the fan is fixed in space and the fan rotation is modeled using grid fluxes. This method is widely used because it provides a fast and effective means of simulating fans. However, the MRF method does not always accurately predict the thermal wake and the mass flow rate through the fan, which causes errors in predicted temperatures on the parts downstream of the fan. Another method for fan simulation is the Rigid Body Motion (RBM) method in which the fan rotates in time. The RBM method models the fan motions directly, thus it can accurately predict the mass flow rate and thermal wake. However, an RBM simulation is transient and needs a time-average to obtain statistically steady-state results. The RBM method requires a significant amount of CPU resources and simulation time, which prevents it from being widely used in industry. In the current work, a Hybrid Rigid Body Motion (HRBM) method is developed and validated. The HRBM method splits the full vehicle thermal simulation into two simulations, and then couples them at the interface. The first simulation is transient, utilizes the RBM method for the fan, and only models the fan regions. The second simulation is steady, which models the full vehicle except the fan regions. The solution from the transient simulation is time-averaged on the exchange interface and used as boundary conditions for the steady simulation. Conversely, the solution for the steady simulation is used as boundary conditions for the transient simulation at the exchange interface. Due to the slight differences resulting from time-averaging, there is a mismatch in the physical quantities at the exchange interface. This causes stability issues which prevent the coupled simulations from converging. Special techniques have been used in this work to stabilize the solution at the interface, which ensured the convergence of the coupled simulations. The HRBM method greatly improves the accuracy of the full vehicle thermal flow simulation compared to using the MRF method. The thermal wake that results from using HRBM to model the fan is very similar to that produced by RBM, but HRBM utilizes ∼20–30% of the simulation resources required by RBM to achieve convergence.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call