论文部分内容阅读
在对超辐射发光二极管(SLD)实施加速退化试验(ADT)并评估其寿命及可靠性的过程中,其性能参数的温度漂移特性影响了评估的精度,为了剔除这种影响,进而提高评估的精度,需要建立相应的温度与性能参数的关系模型来对试验数据进行处理。对SLD的温度建模进行了研究,为了确定一种适用于SLD及ADT数据处理的温度建模方法。通过调研分析讨论了回归方法、人工神经网络、支持向量机等3种温度建模方法;从建模精度、稳定性等方面对上述方法进行了对比分析,并总结了各种温度建模方法的特点;最终确定支持向量机的温度建模方法更适用于开展SLD及ADT的温度建模工作,从而为SLD及相关产品的温度建模和ADT的评估提供了理论依据。
In the accelerated degradation test (ADT) of a superluminescent diode (SLD) and the evaluation of its lifetime and reliability, the temperature drift characteristic of the performance parameter affects the accuracy of the evaluation. In order to eliminate this effect and further improve the evaluation of Accuracy, the need to establish the corresponding temperature and performance parameters of the relationship between the model to test the data processing. The temperature modeling of SLD was studied in order to determine a temperature modeling method suitable for SLD and ADT data processing. Through the research and analysis, three temperature modeling methods, such as regression method, artificial neural network and support vector machine, are discussed and discussed. The methods above are compared and analyzed from the aspects of modeling accuracy and stability, and the methods of temperature modeling are summarized The temperature modeling method of SVM is more suitable for the temperature modeling of SLD and ADT, which provides a theoretical basis for temperature modeling and ADT evaluation of SLD and related products.