Hybrid Virtual Measurement Method for SiC MOSFET Switching Transient Conduction Current Extraction

As a core guarantee for the stable operation of the motor drive systems of electric transportation such as electric vehicle (EV), the accurate measurement of its conduction current provides key data support for the evaluation of the performances of SiC power modules such as switching trajectory optimization, on-line monitoring of junction temperature, and diagnosis of aging status. This paper proposes a new SiC MOSFET conduction current hybrid virtual measurement (HVM) method during switching transient based on second-order passive filtering and deep symbolic optimization (DSO)-based algorithm. Under the high-speed variation of drain current during the switching transient, the Rogowski coil generates an induced potential. The induced voltage can be effectively restored, and the high-frequency oscillation be attenuated by implementing a second-order passive R-C filter. Firstly, this paper proposes and verifies the feasibility of using the filtering time to characterize the switching transient conduction current by analyzing the function of the second-order low-pass filter. Secondly, DSO is proposed to obtain an accurate expression for the drain current in the equivalent current theory. The effectiveness of the proposed method has been validated through real experiments. The results indicate that the filtering time demonstrates high sensitivity and accuracy to the conduction current during switching transient.