Thermal Analysis Of The Asymmetrical Bridge Power Converter In Switched Reluctance Motor System

As a key part of the energy converting, the power converter is widely used in many fields. However, the considerable increase of the power density in the device makes the issue of thermal reliability more prominent. On the other hand, because of many advantages, the SRM(Switched Reluctance Motor) system has been paid high attention since 1980 s. Unfortunately, it is proved that its power converter has high failure rate due to the thermal problem. Therefore, this dissertation studies the thermal analysis of the asymmetrical bridge power converter in SRM system. The main works are generalized as follows:The foundation of this study is the loss calculation of the power converter. Firstly, the steady-state and transient loss models of the power converter are established. In the steady-state model, a piecewise linearized waveform of the phase current is proposed. Then, the transient loss model is built based on the switching loss lookup tables. Furthermore, the influences of system parameters on the device losses are investigated. The results show that: the conduction loss is proportional to the square of the PWM(Pulse Width Modulation) duty cycle, while independent of the PWM frequency, and is proportional to the square of the source voltage, while inversely proportional to the square of the motor speed; the switching loss is proportional to the PWM frequency and the square of the PWM duty cycle; the conduction and switching losses are linear to the turn-on and turn-on angle.As the core of this study, the power converter thermal model is researched. First, the mathematical thermal model is built by the temperature rise, which proves that the temperature rises of the power devices and the heat sink are independent of the ambient temperature. Second, the temperature rise distribution of the power converter is derived from the proposed FEM(Finite Element Method) thermal model. Then, according to the the superposition principle of the heat sink temperature rise with multi heat sources, a novel thermal impedance model is proposed, which can calculate the device transient temperature rise. At last, all the thermal models are verified by the experiment platform by measuring the EMC(Epoxy Molding Compound) temperature. The results show that: the temperatre rise of the chopping transistor is higher than that of the conduction one; the temperature rise curve has a rapid increase when the motor starts working; the rise time is decided by the production of the thermal resistance and the thermal capacity, while independent of the device position and device heating power.To optimize the heat dissipation performance of the power converter, four main aspects of the thermal optimization are investigated by the FEM thermal model, including optimization of the heat sink dimension and heat convection ability, optimization of the control strategy, optimization of the device layout and optimal positions of two rows of devices, optimization of the heat sink placement direction. The optimization results show that: the best way to improve the heat dissipation performance is to increase the heat convection coefficient of the fins; when the devices are placed symmetrically on the heat sink, the thermal reliability of the power converter with alternate PWM control is higher than that with the fixed PWM control; the optimal thermal positions of two devices on the heat sink are only dependent of the heating power ratio between two devices, but independent of the device heating power and the heat sink cooling performance; the power converter has a high thermal reliability with the heat sink placed with the fins opening up.The influence of the ambient relative humidity on the cooling performance of the power converter in mosit air is researched in this chapter. Firstly, the mathematical temperature-humidity model of the power converter in moist air is built based on the heat transfer mechanism of the power converter in moist air. Then, to decouple the temperature-humidity equations, a method treating the moist air as a whole is proposed, and then implemented in ANSYS CFX. Through the FEM model, the simulation results of the device junction temperature rises under different ambient relative humidity and different ambient temperature are derived, which is then validated by the established experiment platform. The results show that: in the case of a fixed environmental temperature, the law of the converter temperature distribution is independent of the relative humidity, while the temperature rises of the power converter decrease as the ambient relative humidity increasing.