摘要
提出并应用了基于格拉布斯准则检验变压器油气监测数据随机误差的自适应处理方法,扭转了缺少自动甄别正态分布下监测数据随机误差手段的被动局面,降低了误告警的概率。首先,对监测数据序列进行分类、排序等预处理,并计算其样本均值、样本标准差;其次,计算序列中最小、最大数据的样本偏离值;再次,基于格拉布斯准则迭代检验各类监测组分的气体,直至排除所有随机误差异常值。实际应用效果表明,该方法改进了监测系统的纠错能力,提升了监测预警的效果。
Study on the adaptive test method for random error of dissolved gas monitoring data in transformer oil, to change the passive situation of the lack of automatic diagnosis of random error means, reduces the probability of false alarm and missing alarm. First, the data classification, sorting, and calculate the average value of the data, the standard deviation. Secondly, the deviation value of the minimum data and the maximum data in the data sequence is solved. Finally, the iterative test monitoring data based on Grubbs criterion, until the exclusion of all outliers. The actual application results show that this method improves the error correction capability of the system, but also enhance the effectiveness of monitoring and early warning.
出处
《电气技术》
2016年第11期91-95,共5页
Electrical Engineering
基金
南方电网公司科技项目(GX2014-2-0025
K-GX2014-020)