摘要
结-环境热网络模型可结合环境温度估计功率器件结温变化,是评估器件散热能力与剩余寿命的重要基础。然而,传统结-环境热网络模型一般假设器件在稳定环境下工作,忽略了环境对流随机性对器件的影响,难以有效反映结温的变化特性。为此,该文提出一种计及环境对流随机性的结-环境热网络模型。该模型首先利用器件历史损耗和温度数据,得到散热器与环境之间的对流热阻样本;再通过小波包变换和马尔可夫链方法,实现对流热阻的随机模拟;最后在指定的电流和环境温度下,模拟器件结温的随机变化。为了验证模型有效性,以功率MOSFET为对象,设计功率器件的随机对流实验平台。实验结果表明,该文所提模型可有效地描述器件结温在随机对流作用下的不确定性变化,能够为器件热安全评估与寿命预测方法发展提供更加真实有效的数据支撑。
Junction-to-ambient thermal network model is an important foundation for evaluating heat-sinking capability and remaining life of power devices,because it can estimate the variation of power device junction temperature with the ambient temperature.However,the traditional junction-to-ambient thermal network model generally assumes that the power devices works in the stable ambient environment,ignoring the random effect of thermal convection on the power device.As a result,it is difficult to effectively describe the variation characteristic of junction temperature.In this paper,a new junction-to-ambient thermal network model of power devices considering the randomness of thermal convective environment was proposed.In the proposed model,the samples of convective thermal resistances between the heat sink and ambient environment were calculated with the historical power loss and temperature data of power device.Then,the wavelet packet transform and Markov chain method were employed to randomly simulate the convective thermal resistance.Finally,the random fluctuation of power device junction temperature can be estimated under the specified current and ambient temperature conditions.A random thermal convection experimental platform was designed and the power MOSFETs were used as the experimental subjects.The experimental results show that the proposed model can effectively describe the uncertain change of power device junction temperature under the action of random thermal convection,and can provide important data support for the development of power device thermal safety evaluation and lifetime prediction methods.
作者
姜鑫
应展烽
万萌
钟震
Jiang Xin;Ying Zhanfeng;Wan Meng;Zhong Zhen(School of Energy and Power Engineering Nanjing University of Science and Technology,Nanjing 210094 China)
出处
《电工技术学报》
EI
CSCD
北大核心
2021年第14期3090-3100,共11页
Transactions of China Electrotechnical Society
基金
国家自然科学基金青年基金(51607091)
智能电网保护和运行控制国家重点实验室开放基金课题资助项目。
关键词
功率器件
热网络模型
随机对流
小波包变换
马尔可夫链
Power device
thermal network model
random thermal convection
wavelet packet transform
Markov chain
