Optimization-Based Parameter Determination for IGBT Models in Network Simulations

In this paper, we propose a novel methodology for determining the model parameters of semiconductor power devices. This method utilizes optimization techniques based on empirical data from actual double-pulse tests, implemented within simulation software to accurately derive model parameters. This approach enables precise replication of the device's switching behavior. Initially, the simulation model for the power semiconductor device and the double-pulse test circuit is constructed using PySpice in a simulation environment. Following this, actual double-pulse tests are conducted to gather extensive waveform data. Finally, these actual switching waveform data serve as the basis for optimizing the model parameters, with the unknown parameters in the model being optimized through iterative methods. This technique not only enhances the accuracy of the device's switching waveforms but also circumvents the need for additional equipment such as curve tracers and high-precision source meters.