IGBT Aging Condition Monitoring And Life Prediction Based On Characteristic Parameters

The insulated gate bipolar transistor(IGBT)is widely used in the power electronic system.If the failure due to aging cannot be detected in advance,it may cause systematic damage and even cause major disasters.This paper mainly studies IGBT aging condition monitoring methods and life prediction methods based on characteristic parameters to achieve real-time assessment of IGBT aging condition,and to issue early warnings before aging failure to improve the reliability of power electronic system operation and the economics of operation and maintenance.On the basis of detailed analysis of the aging failure mechanism of IGBT,for the bonding wire aging,one of the main aging forms,this paper studies the condition monitoring methods based on the conduction voltage drop V_ce and the turn-off process Miller platform time t _GP.In addition to the traditional method based on V_ce at the intersection to eliminate the effect of junction temperature on the monitoring effect,a method based on dynamic adaptive V_ce reference value to eliminate the effect of junction temperature is proposed,which has a wider applicability.In addition,the effectiveness of taking t _GP as a new aging characteristic parameter is proved,and the mechanism of the influence of parameters such as junction temperature on its monitoring effect is analyzed.Based on the comparative evaluation of the above-mentioned condition monitoring methods,the feasibility of applying the method based on the dynamic adaptive V_ce reference value to the IGBT condition monitoring in the traction rectifier is studied.Based on the detailed analysis of the basic structure and control method of the traction rectifier,the corresponding application strategy is proposed,which effectively realizes the monitoring of the aging condition of the IGBT bond wire in the traction rectifier.The power cycle accelerated aging experiment is an important experimental method to study the life of IGBT.Since the size of the power cycle period has a greater impact on the overall experimental efficiency and device aging form,this paper studies the power cycle period adjustment strategy based on multi-parameter optimization.On the basis of detailed analysis of the principles and influencing factors of the temperature rise process and cooling process in the power cycle experiment,it is proposed to adjust the collector conduction current I_c,collector-emitter turn-off voltage V_ce-off,gate drive voltage amplitude V_g,gate resistance R_g,switching frequency f,duty cycle D,heat sink area S,heat sink material,fan power P_w and other parameters at the same time to adjust the power cycle period.This paper also studies the IGBT lifetime prediction method in traction converter based on junction temperature.For IGBT facing complex task conditions in practical applications,the junction temperature fluctuation curve can be obtained by electro-thermal simulation method,and then the temperature load can be extracted by rain current counting method.The analytical life model fitted by the power cycle accelerated aging experiment and linear damage accumulation theory finally estimate the remaining life.In this paper,the IGBT lifetime prediction in subway traction inverter is taken as an example,and the task curve of speed and ramp is considered to calculate and analyze the life.