摘要
应用模糊神经网络预测了电子器件失效阈值随高功率微波参数的变化关系。结合电子器件实验数据较少的情况,提出用可能性理论估计器件失效可能性分布的评估方法;结合模糊神经网络的学习预测能力,得到电子器件失效的可能性分布;并将可能性分布和用信息扩散估计方法得到的概率分布进行了比较,前者能够更好地利用实验数据估计器件失效的可能性。
In this paper, the fuzzy neural network is applied to evaluate the failure thresholds of electronic devices as a function of the parameters of the high power microwave. Based on the characteristic of experimental data, this paper presents a method to evaluate the possibility distribution of electronic device failure applying the possibility theory. Combining the possibility theory and prediction ability of the fuzzy neural network ,the possibility distribution of electronic device failure can be obtained.
出处
《强激光与粒子束》
EI
CAS
CSCD
北大核心
2004年第7期909-914,共6页
High Power Laser and Particle Beams
基金
国家863计划项目资助课题
关键词
模糊神经网络
高功率微波
可能性理论
易损性评估
Fuzzy neural network
High power microwave
Possibility theory
Susceptibility assessment
