融合梅尔频率倒谱系数和本征能量比的风机叶片故障诊断方法  

Fault Diagnosis Method for Wind Turbine Blades Through the Integration of Mel-Frequency Cepstral Coefficients and Intrinsic Energy Ratio

作　　者：何炜 黄帆 何成前 黄卫华[1] HE Wei;HUANG Fan;HE Chengqian;HUANG Weihua(College of information science and engineering,Wuhan University of Science and Technology,Wuhan,Hubei 430081,China;Hunan Hualing Xiangtan Iron and Steel Co.,Ltd.,Xiangtan,Hunan 411104,China)

机构地区：[1]武汉科技大学信息科学与工程学院,湖北武汉430081 [2]湖南华菱湘潭钢铁有限公司,湖南湘潭411104

出　　处：《控制与信息技术》2025年第1期39-43,共5页CONTROL AND INFORMATION TECHNOLOGY

基　　金：国家自然科学基金项目(62303359)。

摘　　要：为解决风机叶片故障诊断的音频信号特征有效提取问题,文章提出了一种融合梅尔频率倒谱系数(Mel-frequency cepstral coefficients,MFCC)和本征能量比(intrinsic energy ratio,IER)的故障诊断方法。其首先分别提取了风机叶片音频信号的MFCC和IER特征,并采用动态时间规整实现故障特征降维,形成复合MFCC特征;然后,对由叶片声脉冲提取的复合MFCC特征进行去噪并采用支持向量机实现对风机叶片故障的分类。实验结果表明,采用基于MFCC与IER的特征提取方法,可实现风机叶片故障识别性能,其正确率达到97.06%,对于风机稳定运行与维护具有重要意义。This paper presents a fault diagnosis method based on Mel-frequency cepstral coefficients(MFCC)and intrinsic energy ratio(IER)to improve efficiency in feature extraction from audio signals for diagnosing faults in wind turbine blades.This methods begins by extracting MFCCs and IERs from audio signals collected from the blades.These fault features are subjected to dimensionality reduction through dynamic time warping,resulting in composite MFCCs.Subsequently,the composite MFCCs,derived from the extraction of acoustic impulses of the blades,are denoised.Faults in the wind turbine blades are then classified utilizing a support vector machine(SVM).Experimental results demonstrated the feasibility and effectiveness of the proposed fault diagnosis method,with a fault recognition accuracy of up to 97.06%,which is of important significance for ensuring the stable operation and maintenance of wind turbines.

关 键 词：风机叶片 故障诊断 梅尔频率倒谱系数 本征能量比 动态时间规整 支持向量机 

分 类 号：TM614[电气工程—电力系统及自动化]