首页|基于气动噪声信号检测模型与无线传感的风电机组叶片故障诊断方法

基于气动噪声信号检测模型与无线传感的风电机组叶片故障诊断方法

Wind Turbine Blade Fault Diagnosis Method Based on Aerodynamic Noise Signal Detection Model and Wireless Sensor

扫码查看
为避免风电机组叶片故障导致的风力发电厂安全运行的问题,提出了基于气动噪声信号检测模型与无线传感的风电机组叶片故障诊断方法,以精准诊断叶片故障.在分析气动噪声信号的基础上,构建基于无线传感的智能监测终端,采集气动噪声信号,将采集信号输入气动噪声信号检测模型中,标准化处理气动噪声信号后,利用ITD方法将信号分解为多个PRC分量,求出气动噪声信号PRC分量能量,重构特征向量并进行PCA降维处理.将降维后的特征向量作为支持向量机数据基础,通过支持向量机对特征向量实施分类,完成风电机组叶片故障诊断.经实验验证:该方法对气动噪声信号特征提取明显,可准确诊断出风电机组叶片故障,分类识别精度高达93%,且诊断结果与实际分类结果基本相同,对风电机组叶片故障诊断的效果较好.
In order to avoid the problem of safe operation of wind power plant caused by wind turbine blade fault,a wind turbine blade fault diagnosis method based on aerodynamic noise signal detection model and wireless sensor is proposed to accurately diagnose the blade fault.On the basis of analyzing the aerodynamic noise signal,an intelligent monitoring terminal based on wireless sensor is constructed to collect the aerodynamic noise signal,input the collected signal into the aerodynam-ic noise signal detection model,standardize the aerodynamic noise signal,decompose the signal into multiple PRC compo-nents using ITD method,calculate the PRC component energy of the aerodynamic noise signal,reconstruct the feature vector and perform PCA dimension reduction processing.The dimensionality reduction feature vector is used as the data basis of support vector machine,and the feature vector is classified by support vector machine to complete the fault diagnosis of wind turbine blades.The experimental results show that the method can extract the characteristics of aerodynamic noise signal ob-viously,and can accurately diagnose the fault of wind turbine blades,with the classification accuracy of 93%,and the diag-nosis result is basically the same as the actual classification result,which has a good effect on the fault diagnosis of wind tur-bine blades.

pneumatic noise signaltest modelwireless sensingwind turbine bladestime-frequency analysisfault di-agnosis

段长江、李发伟、闫文倩、季鹏举

展开 >

陕西吉电能源有限公司,陕西西安 710000

气动噪声信号 检测模型 无线传感 风电机组叶片 时频分析 故障诊断

2024

计算技术与自动化
湖南大学

计算技术与自动化

CSTPCD
影响因子:0.295
ISSN:1003-6199
年,卷(期):2024.43(3)