摘要
分析总结了大量变压器油中溶解气体含量的历史数据,探究了气体产生与变压器故障之间的内部联系,并结合人工智能方法构建了变压器故障智能诊断模型,模型以支持向量机为基础,以麻雀搜索算法寻解,引入反向学习策略和萤火虫干扰策略进一步提升寻优能力。仿真试验分析结果表明,该模型能够迅速收敛,对变压器故障诊断的准确率可达96%,高于现有智能诊断方法。
By analyzing and summarizing the massive historical data on the content of dissolved gas in transformer oil,the internal relationship between gas generation and transformer faults is studied,based on which an intelligent diagnosis model for transformer faults is built combined with the artificial intelligence technique.The model,based on support vector machine,uses the sparrow search algorithm for a solution,and introduces the reverse learning strategy and the firefly interference strategy to further enhance its optimization capacity.The analysis on simulation tests show that the model is available to converge rapidly and its diagnosis accuracy rate for transformer faults reaches up to 96%,higher than that based on the existing intelligent diagnosis methods.
作者
董瑞钧
DONG Ruijun(Power China Jiangxi Electric Power Engineering Co.,Ltd.,Nanchang 330096)
出处
《电力安全技术》
2025年第1期39-43,共5页
Electric Safety Technology
关键词
变压器
溶解气体特征
支持向量机算法
麻雀搜索算法
智能诊断
transformer
dissolved gas characteristic
support vector machine algorithm
sparrow search algorithm
intelligent diagnosis
