摘要
电容型设备介质损耗因数(tanδ)的在线监测会受到环境因素(温度、湿度、污秽等)的影响,对其进行合理修正非常必要。根据人工气候室内变压器套管tanδ的在线测量试验结果,分析了环境因素对电容型设备tanδ的影响情况,提出了基于最小二乘支持向量机的主要环境因素对tanδ影响的修正模型,并采用遗传算法优化了支持向量机的参数。该模型可用于将非标准大气条件下电容型设备的tanδ测量值修正为标准大气条件下的值,排除环境因素影响,增加在线监测结果的可比性,实际的修正结果验证了该方法的有效性。
Precision of on-line dielectric loss factor （tanδ） monitoring can be affected by environmental factors, so it is necessary to modify it. According to the tanδ test results of transformer bushings in the artificial climate chamber, the effect of environmental factors on the tanδ of capacitive equipment was analyzed. Then based on the least squares support vector machine （LS-SVM） method, the modificatory model of the main environmental factors effect on the tanδ was advanced, and the genetic algorithm （GA） method was used to optimize the parameters of this model. This model can be used to modify the tanδ of capicitive equipment from nonstandard atmosphere to standard atmosphere, and to get rid of the influence of environmental factors. So that the comparability of on-line tanδ monitoring value is improved. The effectiveness of this method is validated by calculation results of practical modificatory examples.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2009年第4期123-128,共6页
Proceedings of the CSEE
关键词
介质损耗因数
支持向量机
遗传算法
修正
dielectric loss factor
support vector machine
genetic algorithm
modification
