A fast junction temperature prediction method for asymmetric half-bridge power converters

Abstract

A method for predicting the junction temperature of three-dimensional resistance-capacitance thermal network of asymmetric half-bridge power convert features that the junction temperature of the three-dimensional resistance-capacitance thermal network is predicted by using the junction temperature of the three-dimensional resistance-capacitance thermal network of asymmetric half-bridge power converter. In this method, the junction temperature of power devices in asymmetric half-bridge power converters under different loads is predicted. The junction temperature coupling and boundary conditiondecoupling are taken into account, and a three-dimensional resistance-capacitance thermal network model is established to realize the effective junction temperature prediction of power devices. Firstly, the transient loss distributions under different load conditions are obtained in asymmetric half-bridge power converters. Secondly, the finite element model of asymmetric half-bridge power converter is established, and the coupling thermal impedance and boundary decoupling thermal impedance are extracted according to the finite element thermal stress simulation with the transient loss as the input of thermal load. Then, the expressions of coupled thermal impedance and heat dissipation boundary conditions are obtained by means of electrothermal analogy theory and curve fitting. Finally, a three-dimensional resistive-capacitive thermal network model is constructed to predict the junction temperature of the power device by using the controlled voltage source to represent the thermal coupling impedance relationship and the sub-radiator thermal impedance to represent the different thermal boundary conditions.

Description

technical field [0001] The invention relates to a method for predicting the junction temperature of a power device of an asymmetrical half-bridge power converter of a switched reluctance motor, and is especially suitable for asymmetrical half-bridge power converters with different phase numbers and different power levels. Background technique [0002] Prediction of junction temperature of power devices in asymmetrical half-bridge power converters is the basis for reliable operation of switched reluctance motor systems. Usually, the junction temperature of the power device can be obtained by establishing the finite element thermal model of the converter, but the unreasonable solution time of the finite element model will seriously reduce the prediction speed of the junction temperature, especially in the field of reliability-oriented optimization . Although the one-dimensional RC network represented by the Foster model and the Cauer model can quickly obtain the junction temp...

AI Technical Summary

Problems solved by technology

Usually, the junction temperature of the power device can be obtained by establishing the finite element thermal model of the converter, but the unreasonable solution time of the finite element model will seriously reduce the prediction speed of the junction temperature, especially in the field of reliability-oriented optimization

Although the one-dimensional RC network represented by the Foster model and the Cauer model can quickly obtain the junction temperature, the one-dimensional RC network is only suitable for a single device or a power converter with a small coupling.

Due to the particularity of the asymmetrical half-bridge power converter topology of switched reluctance motors, there are almost no integrated converter modules. Therefore, separate MOSFETs and diodes are usually used to build converters, which leads to thermal coupling of power devices through radiators. Rather than being thermally coupled inside the power device, there will be strong thermal coupling

At the same time, due to the unique control method of the switched reluctance motor, the same type of devices in the same phase will bear different electrical stresses, which further aggravates the imbalance of heat distribution

Images