A MOSFET EMC modeling method based on electrical characteristic measurement and simplex optimization and particle swarm optimization

Metal oxide semiconductor field effect transistors (MOSFETs) are widely used in various power electronic systems, and the establishment of the electromagnetic compatibility (EMC) model for MOSFETs is crucial for EMC analysis of these systems, especially high-frequency switching circuits. In this paper, a MOSFET EMC modeling method is proposed based on electrical characteristic measurement, simplex optimization, and particle swarm optimization (PSO), to make MOSFET models meet both functionality and EMC analysis requirements. First, systematic methods for extracting SPICE parameters based on physical MOSFETs are presented, including a proposed curve-fitting method based on PSO for extracting the CDS-VDS and ID-VGS characteristic parameters. Second, parametric influence analyses are conducted on some important model parameters, and the law that these parameters affect the EMC of MOSFETs is obtained. Finally, the preliminary MOSFET model was optimized comprehensively using the simplex method. The measured and simulated results are in good agreement in both time domain and frequency domain. It proves that the established model meets the requirements for model accuracy in terms of both EMC and functionality. Therefore, the proposed EMC modeling method is feasible. This paper proposes a systematic approach for establishing electromagnetic compatibility (EMC) models of metal oxide semiconductor field effect transistors (MOSFETs). The curve-fitting method based on particle swarm optimization is proposed to address the non-linear curve fitting problem with multiple local optimal solutions. The simplex method is used to comprehensively optimize the preliminary model through the printed circuit board (PCB)-level EMC simulation.image