基于多种人工智能技术集成的电力变压器故障诊断研究

Power Transformer Fault Diagnosis Research Based on Integration of Multiple Artificial Intelligence Technologies

这篇文章会针对许多个AI技术集合而成的电力变压器出现问题从而开展判断确诊的有关研究。这篇文章涉及的问题归纳了演化算法、Fuzzy Logic、ANNs、案例推导等等几种AI技术的独到之处,填补了独个AI技术的弱点,减少了数据库范本的信息差别,提高了信息初始权值,很大程度上升了ANNs的约束性。ANNs的网络结构以及测试范本经过很多的测算度量而后挑选认定,并经过在于案例推导的专业软件检测到了相对优等的资源范例并总结了判断确诊的定论。确诊成果显示,这个诊断手段可以比较精确判断区别出电力transf.(T)的问题所在。

The paper researches of diagnosis and confirmation of problems of multiple AI power transformers,sum up unique features of several AI technologies in terms of problems,including evolutionary algorithm,Fuzzy Logic,ANNs,case derivation,etc.,fills in weakness of single AI technology,reduces information difference of database template,improves initial weight of information,and greatly increases constraint of ANNs.After calculation and measurement,selection and identification,and test with professional software based on case derivation,relatively superior resource examples and diagnosis and confirmation have been concluded for network structure and test model of ANNs.Diagnosis results show the diagnosis method can accurately distinguish problem of transf(T).