Modelling and Validation of a Switched Reluctance Motor Stator Tooth with Direct Coil Cooling

Abstract

This paper presents the modelling and validation of an advanced thermal lumped parameter (LP) model for a stator tooth of a switched reluctance motor (SRM) with a dry lateral slot cooling method. Standard and simple lumped parameter models for electric motors can insufficiently predict the temperature distribution within the components of the motor. In standard LP models, only several nodes are used to model each component, while more accurate models are needed to predict the effect of different cooling methods on the thermal performance of the motor without the need for experiments. A fully 3D thermal finite element (FE) model could be used but this would increase effort, complexity and computing time unnecessarily. Therefore, an advanced 3D LP model including the dry lateral slot cooling method was developed and validated based on experiments on a real stator tooth cooled with the modelled cooling method. The 3D LP model is extracted from a 2D FE radial simulation of the stator tooth and extended axially in 3D to include axial heat transfer. Experiments were performed with a setup consisting of one tooth of a SRM without rotor, but including stator iron, one winding and two triangular stainless steel tubes in the slots at both sides of the winding cooled by a 60/40% mixture by mass of water-glycol. The setup is equipped with several thermocouples integrated within the components to determine the component temperatures. Three inlet temperatures (20, 35 and 50°C) and four flow rates (2, 6, 9 and 13 l/min) of the coolant were tested at three different heat losses in the winding (10, 30 and 50 W). A comparison between the simulated and measured temperatures showed generally higher temperatures in the experiment. The presence of imperfections in the manufacturing of the experimental setup was determined as the cause of this offset. These imperfections result in lower material thermal conductivities and higher contact resistances than expected from scientific literature. After fitting those thermal properties on the measurements, similar simulated temperatures could be obtained as in the experiments.

Quality Assessment of a 2D FE Based Lumped Parameter Electric Motor Thermal Model Using 3D FE Models

“…Experimental measurements or other numerical models can be used as a reference to verify the developed model. Possible issues with experimental measurements are the accuracy of the measurements within electric machines and uncertainty of the material properties and thermal contact resistances of components [12]. To exclude these uncertainties within the model verification, another reliable numerical model will be used to simulate the 3D conduction within the motor test case.…”

“…This can be done by comparing the simulation results with experimental measurements or other numerical models. To exclude possible issues with the accuracy of the measurements within electric machines and uncertainty of the material properties and thermal contact resistances of components [12], it was decided to verify the 3D LPM results with the results of a 3D FE simulation in OpenFOAM based on exactly the same inputs.…”

Comparison of Synchronous Reluctance Motors and Permanent Magnet Synchronous Motors with Direct Liquid Cooling Arrangement

“…In the latest papers related to the research of relatively compact motor drive systems (in the range of 1 MW) , DLC is applied only to Permanent Magnet motors [9]- [10] and to Switched reluctance motors [11]- [12]; a particular case of DLC is explained and discussed in [13] in which the windings are directly cooled in a slotless motor.…”

“…In [11]- [12], the DLC is applied to a Switched Reluctance Motor. Differently from the previous case, the cooling tubes are in the slot and not in the wires; therefore, the cooling capabilities of the system could be lower compared with DLC integrated into the wires.…”