Temperature Estimation Based on Parameter Identification in Electric Motor Drive Using a Single DC-Link Current Sensor

This paper proposes a method to estimate the stator and rotor temperatures of a permanent magnet (PM) motor using a single DC-link current sensor (SDC-LCS). This approach eliminates the need for three-phase current sensors and temperature sensors, thereby reducing the system's price and complexity. A current reconstruction technique is introduced using the SDC-LCS without compromising control performance. Based on the reconstructed currents, a partial sampling method is used to estimate temperature-related parameters. Subsequently, a ridge regression model is trained to predict the stator and rotor temperatures from these parameters. The effectiveness of the proposed estimation framework is experimentally verified using a 120 W motor under various operating points.