Study On Physical Modeling Of IGBT And Its Drive

Insulated gate bipolar transistor(IGBT) is equivalent to a giant transistor(GTR) driven by metal oxide semiconductor field effect transistor(MOSFET) on the physical structure. Therefore, it has the same advantages with MOSFET and GTR, such as high input impedance, low conductance voltage drop, simple drive circuit and so on. IGBT has been widely used in the field of power electronics, due to its high switching frequency and large output power. However, its switching loss, improper use, and failure condition are closely related to efficiency and reliability of the power electronic system. Therefore, the physical model of IGBT is researched in this paper. The structural parameters of the specialized IGBT model are used to characterize the difference between devices. And the drive is analyzed and optimized based on physical model to improve the efficiency of the devices and the reliability of system.In this paper, the structure characteristics and working mechanism of IGBT devices are introduced, while the switching process and the output characteristics are described by using the motion of electrons and holes. By analyzing the two main modeling methods of IGBT devices, the physical modeling method is taken as the research content of the subject.Based on the analysis and derivation of Hefner model of IGBT device, the physical model of IGBT is improved.The main method is establishing turn-on and turn-off model respectively to revise the shortcoming of Hefner's model whose output leads the experimental at the transient of turnig off. The correctness and accuracy of the improved model are verified by the comparison between the output of simulation model and an actual IGBT device. Meanwhile, the difference between devices is characterized by using structural parameters. The closed-loop identification method of model parameters based on neural network optimization algorithm is presented to obtain the IGBT model. This model has the significance of characterizing the difference between devices. The simulation results show that the improved model can achieve higher accuracy and accurate prediction of IGBT devices' output.The drive circuit of an IGBT is an important factor to affect its output characteristics. Based on IGBT physical model, the corresponding optimization strategies are proposed for the two different optimization objectives respectively in this paper. One is to solve the inconsistency problem of devices' output caused by the inherent differences between the same type devices. The influence relationship between model parameters and device output and the specific content of drive circuit model are discussed. The method is changing external characteristics, namely the driving resistance, electrode capacitance and parasitic inductance value according to the change direction of model parameters values to realize the consistency of IGBT devices' output. The other one is to realize the quick and reliable output of IGBT device, and the proposed method is taking use of the proportional-integral control(PI control) with ideal output of IGBT device as its control setting values to implement closed-loop control for IGBT device model. Finally, the feasibility of the two optimization schemes is verified in MATLAB/simulation platform. The results show that the rapid and reliable switching of IGBT device and the decrease of switching loss are realized.