Search results

For an accurate simulation of the temperature distribution inside an electrical machine a method for deriving the convective heat transfer coefficient numerically would be desirable. The purpose of this paper is to present a reliable simulation setup, which is able to reproduce the measured convective heat transfer coefficient at certain spots on the end windings of an electric machine.

The heat flux density on certain spots on the end windings of an induction motor have been measured with heat flux sensors, in order to find out the convective heat transfer coefficient. To identify the air mass flow inside a cooling duct of an encapsulated cooling circuit during the operation of the motor, the pressure loss inside the duct has been measured. The measured data for temperature and air mass flow have been used as boundary conditions for the identification of the convective heat transfer coefficient with a commercial software for computational fluid dynamics (CFD).

The measured data for the local convective heat transfer coefficients have been compared to the results of the numerical simulation for various rotational velocities. The quality of the simulated convective heat transfer coefficient depending on the rotational velocity meets the measured values. Owing to the used simplified model, the quantity of the measured values differ strongly around the simulated coefficient for the convective heat transfer.

The derivation of the convective heat transfer is a challenging subject in CFD but has become more reliable with the invention of the SST and the SAS‐SST turbulence model. In the present work, measurements on the end windings have been compared to simulation results derived with the SAS‐SST turbulence model.

Convective heat transfer features of a turbulent slot jet impingement are comprehensively studied using two different computational approaches, namely, URANS (unsteady Reynolds-averaged Navier–Stokes equations) and SAS (scale-adaptive simulation). Turbulent slot jet impingement heat transfer is used where a considerable heat transfer enhancement is required, and computationally, it is a quite challenging flow configuration.

Customized OpenFOAM 4.1, an open-access computational fluid dynamics (CFD) code, is used for SAS (SST-SAS k-ω) and URANS (standard k-ε and SST k-ω) computations. A low-Re version of the standard k-ε model is used, and other models are formulated for good wall-refined calculations. Three turbulence models are formulated in OpenFOAM 4.1 with second-order accurate discretization schemes.

It is observed that the profiles of the streamwise turbulence are under-predicted at all the streamwise locations by SST k-ω and SST SAS k-ω models, but follow similar trends as in the reported results. The standard k-ε model shows improvements in the predictions of the streamwise turbulence and mean streamwise velocity profiles in the zone of outer wall jet. Computed profiles of Nusselt number by SST k-ω and SST-SAS k-ω models are nearly identical and match well with the reported experimental results. However, the standard k-ε model does not provide a reasonable profile or quantification of the local Nusselt number.

Hybrid turbulence model is suitable for efficient CFD computations for the complex flow problems. This paper deals with a detailed comparison of the SAS model with URANS and LES for the first time in the literature. A thorough assessment of the computations is performed against the results reported using experimental and large eddy simulations techniques followed by a detailed discussion on flow physics. The present results are beneficial for scientists working with hybrid turbulence models and in industries working with high-efficiency cooling/heating system computations.

The purpose of this paper is to study the pressure fluctuation characteristics in the sidewall gaps of a centrifugal dredging pump in detail and discover the excitation sources.

An unsteady numerical simulation with shear–stress transport–scale-adaptive simulation (SAS-SST) model was conducted for a centrifugal pump considering the sidewall gaps. The numerical codes were validated by a model test carried out in China Water Resources Beifang Investigation, Design and Research Co., Ltd. Fast Fourier transform was used to obtain the frequency components of the pressure fluctuation.

Pressure fluctuation characteristics inside the pump were analyzed for a condition near the design point. In the sidewall gaps, the circumferential, radial and axial distribution of the pressure fluctuation amplitude follow different laws. The non-axisymmetrical distribution of pressure fluctuation in the sidewall gaps shows that the unsteady flow in the volute casing which has a non-axisymmetrical geometry imposes an evident effect on the flow field in the sidewall gaps and the interaction between the main flow and the clearance flow cannot be neglected. There are several frequency components appearing as the dominant frequencies at different locations in the sidewall gaps, but the relatively stronger pressure fluctuations are all dominated by the rotating frequency. It indicates that the rotating impeller, which originally makes the shrouds rotate, is the primarily excitation source of the pressure fluctuations in the sidewall gaps.

The pressure fluctuation characteristics in the sidewall gaps of centrifugal pumps were first comprehensively analyzed. Unsteady flows in the sidewall gaps should be considered during the design and operation of centrifugal pumps.