摘要
35 kV箱式变压器的安全稳定运行对配电网可靠性至关重要,然而其面临着发热故障的隐患。提出了一种基于计算机通信的发热故障诊断方法。该方法利用光纤光栅传感器进行温度数据采集,通过RS-485和工业以太网实现数据传输,并结合支持向量机(Support Vector Machine,SVM)和卷积神经网络(Convolutional Neural Network,CNN)算法进行故障智能诊断。实验结果表明,与传统方法相比,设计方法在故障检测率、误报率、漏报率以及诊断速度等方面均有显著提升,为变压器智能化运维提供了新的技术手段。
The safe and stable operation of 35 kV box type transformers is crucial for the reliability of distribution networks,but they face the hidden danger of heating faults.The article proposes a method for diagnosing heating faults based on computer communication.This method utilizes fiber Bragg grating sensors for temperature data acquisition,achieves data transmission through RS-485 and industrial Ethernet,and combines Support Vector Machine(SVM)and Convolutional Neural Network(CNN)algorithms for intelligent fault diagnosis.The experimental results show that compared with traditional methods,the proposed method significantly improves the fault detection rate,false alarm rate,false alarm rate,and diagnostic speed,providing a new technical means for intelligent operation and maintenance of transformers.
作者
李金秀
LI Jinxiu(Southwest Branch of China Electric Power Construction New Energy Group Co.,Ltd.,Chengdu 610000,China)
出处
《通信电源技术》
2024年第24期37-39,共3页
Telecom Power Technology
关键词
箱式变压器
发热故障
计算机通信
智能诊断
box-type transformer
heating failure
computer communication
intelligent diagnosis
