Search

IGBT power module is the key component of the power electronic converter, but it has the lowest reliability. The junction temperature is the crucial factor which affects power module’s reliability. To some extent, the power handling capability of the converter depends on the thermal stress of the power module. Thermal management is an effective method to improve the reliability of power device, as well as enhance the power capability. For this purpose, this paper introduces the reliability design to the power converter’s traditional compensation controller design for the first time. A new concept of generalized dual-loop controller, which includes temperature control loop and electric power control loop, is proposed. The reliability and stability of the system are both considered, with the help of the hybrid controller, the power converter can operate steadily with higher reliability. The novelty of this paper is to improve the thermal control method of carrier frequency adjustment through experimental implementation during the full life cycle of the converter. The target is to control the temperature variation to be almost a constant value as well as extend the lifetime of the converter. IR sensor is used to measure the chip temperature of the unpackaged IGBT module. The temperature variation and the average temperature are all considered in thermal management, from the reliability improvement point of view. At last, the idea is digital implemented based on a varying load of power inverter system with real-time measurement of the chip’s surface temperature. [ABSTRACT FROM AUTHOR]

Power semiconductors in the wind turbine power converter system suffer from two-scale thermal loadings, the fundamental frequency thermal cycling caused by the output frequency of converter and the low frequency thermal cycling due to the variation of long-term wind speed. These two-scale thermal loadings introduce different consumed lifetimes. Accurate lifetime estimation in the wind power application is desired for reliability prediction and health management. This paper adopts the Bayerer lifetime model to evaluate the consumed lifetime of power semiconductors in wind power converter systems based on a numerical junction temperature calculation method. Lifetime estimation can be improved by taking into account the ambient temperature. Studies show that fluctuations of the ambient temperature increase the consumed lifetime due to the low frequency thermal cycling, but have little effect on the consumed lifetime due to the fundamental frequency thermal cycling. Our results also show that the consumed lifetime due to fundamental frequency thermal cycling mainly falls on the high wind speed area, whereas the consumed lifetime due to low frequency thermal cycling is clustered in the area due to large low frequency junction temperature fluctuations. The resulting distribution characteristics can be used in the thermal management for reliability improvement. [ABSTRACT FROM AUTHOR]