摘要
目前潞安化工集团下属各大煤矿井下电力变压器故障特征向量均无法全面反映变压器故障问题,因此无法在最短时间内快速定位故障原因,为了提高电力变压器故障检测的准确度,现设计了一种基于改进鲸鱼算法优化支持向量机的电力变压器故障检测方法。该方法通过收集电力变压器正常运行及出现故障时的关键数据,按照检测的顺序将样本进行分类处理,选取电力变压器故障检测特征量,建立基于鲸鱼算法的预测模型,并采用支持向量机算法对故障进行检测,以此在最短时间内查找出电力变压器真正的故障原因,减少停电对井下安全生产造成的不利影响。实验结果表明,在几个变压器故障类型的识别上,该方法都具有较高的识别率和准确率,完全可以在井下实际故障处理中进行应用。
Currently,the fault feature vectors of underground power transformers in major coal mines under Lu'an Chemical Group cannot fully reflect the problem of transformer faults,so it is impossible to quickly locate the cause of faults.In order to improve the accuracy of power transformer fault detection,a power transformer fault detection method based on improved whale algorithm optimized support vector machine is designed.This method collects key data on the normal operation and occurrence of faults in power transformers,classifies and processes the samples in the order of detection,selects fault detection feature quantities for power transformers,establishes a prediction model based on the Whale Algorithm,and uses support vector machine algorithm to detect faults,in order to identify the true cause of power transformer faults in the shortest possible time and reduce the adverse impact of power outages on underground safety production.The experimental results show that this method has high recognition rate and accuracy in identifying several types of transformer faults,and can be fully applied in practical fault handling underground.
作者
郑云龙
吴超
ZHENG Yunlong;WU Chao(Changcun Coal Mine of Shanxi Lu'an Chemical Group 046000,Changzhi City,Shanxi Province;School of Electrical and Control Engineering,Liaoning University of Engineering and Technology 123000,Fuxin City,Liaoning Province)
出处
《煤矿现代化》
2025年第2期86-90,共5页
Coal Mine Modernization
关键词
改进鲸鱼算法
支持向量机
电力变压器
故障检测
improving whale algorithm
Support Vector Machine
Power transformer
Fault detection
