摘要
综合人耳听觉模型及小波变换多分辨率特性进行电气设备异响信号特征分析,提出了一种符合变压器运行声音信号功率谱特征的多分辨率高斯滤波器组时频分析方法。通过对人耳听觉系统非线性特性,设计高斯滤波器组对数变化中心频率,同时基于小波变换思想扩展了滤波器组的多分辨率特性。详细讨论了基于多分辨率高斯滤波器组功率响应及临界带宽选取的滤波器参数设计方法。分析了多分辨率高斯滤波器组相比等带宽高斯滤波器组的多分辨率特性及滤波器组倒谱系数特征识别性能。
Based on the analysis of abnormal signal characteristics of electrical equipment by synthetic human auditory model and multi-resolution characteristics of wavelet transform,a multi-resolution Gauss filter bank time-frequency analysis method accord with the power spectrum characteristics of transformer running sounding signals is proposed. Based on the nonlinear characteristics of the human auditory system,the center frequency of the gauss filter bank is designed,and the multi-resolution characteristics of the filter bank are extended based on the wavelet transform. The filter parameter design method based on the power response and bandwidth selection of multi resolution gauss filter bank is discussed in detail.The multi-resolution characteristics and the characteristics of filter bank cepstrum coefficients of multi-resolution gauss filter bank compared to the equal bandwidth gauss filter banks are analyzed in this paper.
出处
《中国电子科学研究院学报》
北大核心
2017年第6期654-661,共8页
Journal of China Academy of Electronics and Information Technology
基金
国网陕西省电力公司科学技术项目(5226KY140017)
关键词
高斯滤波器组
故障诊断
时频分析
多分辨率
听觉模型
Gauss filter bank
fault diagnosis
Time frequency analysis
Multi resolution
Auditory model
