【摘要】 SiC功率管器件广泛应用在航空领域的电能变换、配电等场合,其健康状况十分重要。在SiC器件的健康监测应用中,导通电阻的检测是一项十分重要的技术。为了能够简单准确地得到碳化硅(SiC)MOSFET功率器件的导通电阻,本文提出了一种基于神经网络的SiC MOSFET器件导通电阻估测方法。本文搭建SiC MOSFET导通电阻测试电路仿真和物理试验平台,并使用BP神经网络(BP neural networks, BPNN)对不同温度、不同栅极电压以及不同漏极电流下SiC MOSFET器件的导通电阻数据进行详细描述。最后,对基于BPNN的SiC MOSFET导通电阻估测方法进行效果验证。结果表明,该方法具有精度高和泛化能力强的优点,能够实现SiC MOSFET器件导通电阻的有效估测。更多还原

【Abstract】 SiC power devices are applied widely in the applications of aeronautics, mainly in terms of electrical energy transform and distribution. The health surveillance technology has been a focus in the past time for these devices,whose on-state resistance detection technology is crucial. In order to obtain on-state resistance of silicon carbide(SiC) MOSFET power devices simply and accurately, this paper proposes a neural network based on-state resistance estimation method for SiC MOSFET devices. This paper builds a simulation and physical experiment platform for SiC MOSFET on-resistance test circuit, and uses BP neural network to describe in detail on-stateresistance data of SiC MOSFET devices under different temperatures, different gate voltages and different drain currents. Finally, the effect of SiC MOSFET on-state resistance estimation method based on BP neural network is verified. The results show that this method has the advantages of high accuracy and strong generalization ability, and can realize effective estimation of on-state resistance of SiC MOSFET devices.更多还原