Rotor temperature estimation of squirrel-cage induction motors by means of a combined scheme of parameter estimation and a thermal equivalent model

This paper deals with a rotor temperature estimation scheme for fan-cooled mains-fed squirrel-cage induction motors. The proposed technique combines a rotor resistance estimation method with a thermal equivalent circuit. Usually, rotor resistance estimation works quite well under rated load conditions. By contrast, if the motor is slightly loaded, rotor resistance estimation becomes inaccurate due to the small slip. Therefore, rotor temperature estimation under low-load conditions may be estimated by a thermal equivalent model. In order to determine the rotor resistance and, thus, rotor temperature accurately, several machine parameters have to be obtained in advance. Load tests provide the leakage reactance and the iron losses of the induct machine. The stator resistance has to be measured separately. The parameters of the thermal equivalent model are a thermal resistance and a thermal capacitance. These parameters are derived from a heating test, where the reference temperature is provided from the parameter model in the time domain. This lumped thermal parameter model is based on the assumption that the total rotor temperature increase is caused by the total sum of the losses in the induction machine. Measuring results of a 1.5-kW and an 18.5-kW four-pole low-voltage motor and a 210-kW four-pole high-voltage motor are presented and compared.