Advanced Data-based Temperature Modeling of an External Rotor PM Machine Considering an Optional Housing

The focus of this study is on the transient thermal behavior of a brushless PMSM with external rotor and an optional housing. Therefore, a two-stage modeling process consisting of a classical thermal modeling using lumped parameter thermal networks (LPTNs) followed by a data-based parameter modification y ields a t hermal model with v ery high prediction accuracy across the entire safety operating area (SOA). First, the initial parameter determination is based on analytical and empirical approaches. Then, a generic algorithm is used to compensate uncertainties caused by model simplifications and a pproximations. For t his r eason, a f ully automated data acquisition i s r equired. A t est bench that allows a simultaneous acquisition of mechanical, electrical, and temperature data is discussed. Finally, the developed models are verified using measured data.