Based On Finite Element Method And Electrothermal Multiphysical Field Simulation And Thermal Stress Analysis Of IGBT

With the increasing consumption of coal,natural gas and other resources in the world,the development of energy technology has been paid more and more attention by the public,and the performance requirements of power electronic equipment are getting higher and higher.How to conduct effective state monitoring for IGBT to ensure its reliability is becoming more and more important as a core component of power electronic equipment.Because the mechanical stress waves reflecting the state and fault are generated by the coupling of multiple physical fields such as electromagnetic and thermodynamic fields in IGBT devices under switching and fault states,so stress wave detection method can be used as a potential IGBT state monitoring scheme.Compared with the traditional state monitoring methods of electric,thermal and magnetic parameters,the state monitoring based on mechanical stress wave parameters is a new online monitoring method,which shows the advantages of rapid-ness,nonintrusive and accuracy.The establishment of state monitoring method based on mechanical stress wave is of great significance to improve the reliability of devices and reduce economic losses.Regrettably,since 2014,relevant scholars have paid attention to the mechanical stress wave of power module,but until now,the research on the mechanism and propagation mode of mechanical stress wave has not been indepth at home and abroad.In order to explore the mechanism and influencing factors of internal mechanical stress wave in IGBT,this thesis uses finite element analysis software to build an electric-thermal-mechanical multi-physical field platform to conduct steady state and transient analysis.According to the maximum mechanical stress under the coupling of multiple physical fields and the influence of each structure on the mechanical stress,the causes of IGBT stress wave in switching state and the factors affecting the stress wave signal are studied,and according to the calculated results of the maximum stress,some suggestions are provided for the structure optimization of IGBT modules.Firstly,the effects of drift layer thickness and drift layer concentration on the cellular characteristics of IGBT are studied by TCAD software to provide suggestions for improving the device performance.The cell structure is optimized to obtain the parameters such as the conductivity of the chip,the switching duration of one cycle and the current change curve under the cycle,which provided accurate input parameters for the IGBT model established under ANSYS analysis software in the next step.Sencondly,the steady-state multi-physical field simulation of IGBT module is carried out by ANSYS,and the switching characteristics obtained from TCAD are used as input excitation in chapter 4.The simulation results show that the thermal stress is the main factor affecting the mechanical stress wave.The simulation analyses the influence of substrate material and thickness,solder layer material and thickness,liner material and thickness on the IGBT module stress,and provides suggestions for the structure optimization of IGBT module.Finally,chapter 5 simulation the temperature response and stress response of different structural layers,as well as the displacement,velocity and acceleration of the solder layer surface in the normal direction of the time-history diagram,with the material and thickness of the solder layer,under different working currents,the fracture of the bond line,the solder layer under the condition of cavity and crack.The view that thermal-mechanical stress can produce stress wave is verified by simulation,and it can propagate between structures.By analyzing the similarities and differences of stress response time domain and frequency domain curves under normal state,different composition materials and thicknesses,different working current and different degree of aging,it is found that current and degree of aging will affect stress wave time domain and frequency domain parameters.This thesis preliminarily proves that the stress wave method has a good research prospect for state monitoring of IGBT modules.