摘要
为提高变压器运维的智能化水平,提出一种基于声纹压缩的变压器状态检测新方法。该方法利用声纹特征提取技术从变压器运行声音中提取关键特征,应用声纹压缩算法实现高效编码,并基于压缩后的特征训练状态识别模型。实验结果表明,该方法在故障检测率、虚警率、准确率等指标上均优于传统方法,可有效提升变压器状态检测的效率和可靠性。
To improve the intelligence level of transformer operation and maintenance,a new method for transformer state detection based on voiceprint compression is proposed.This method utilizes voiceprint feature extraction technology to extract key features from the sound of transformer operation,applies voiceprint compression algorithm to achieve efficient encoding,and trains a state recognition model based on the compressed features.The experimental results show that this method outperforms traditional methods in terms of fault detection rate,false alarm rate,accuracy,and other indicators,effectively improving the efficiency and reliability of transformer statedetection.
作者
李敏昱
程航
黄宇轩
蔡嘉炜
高瑞鑫
孙斌
LI Minyu;CHENG Hang;HUANG Yuxuan;CAI Jiawei;GAO Ruixin;SUN Bin(Fuzhou Power Supply Company of State Grid Fujian Electric Power Co.,Ltd.,Fuzhou 350000,China)
出处
《电声技术》
2024年第10期48-50,共3页
Audio Engineering
基金
国网福建省电力有限公司科技项目(521310230009)。
关键词
变压器状态检测
声纹压缩
特征提取
transformer status detection
voiceprint compression
feature extraction
作者简介
李敏昱(1988-),男,硕士,高级工程师,研究方向为配电运检。
