摘要
为满足风电机组处理故障数据准确性和实时性的要求。文章通过采集无线风电机组振动信号,对其进行数学建模,利用小波分析提取振动信号的随机噪声和状态信号叠加,并以此为观测方程。利用小波包分解求取降噪前和降噪后的信号,根据各个频带能量变化提取故障信号,并采用SVM方法进行故障模式识别,从而实现对风电机组的故障定位。实验验证了该算法能有效提高风电机组故障定位的精确性和可靠性。
For the requirement of wind turbines real time and accuracy of fault data,the wind turbine vibration signals are collected.Then mathematical modeling of the collected vibration signals is setup as a state equation,and wavelet analysis is used to extract the vibration signal of random noise and state signal superposition as observation equation.Wavelet packet decomposition for noise reduction before and after noise signal,according to the different frequency band energy change is to extract the fault signal,and the SVM method is used to identify the fault mode,so as to realize fault location for wind turbines.Experiment shows that this algorithm can effectively improve the accuracy and reliability of wind power unit fault location,the stable operation of wind power units have very strong application value.
作者
马小娟
马常胜
Ma Xiaojuan Ma Changsheng(Tianjin Transportation Vocational College, Tianjin 300110, China Tianjin Jin＇an Thermal Power Co.,Ltd, Tianjin 300204, China)
出处
《可再生能源》
CAS
北大核心
2017年第9期1341-1346,共6页
Renewable Energy Resources
基金
天津市科技计划项目(12ZCDGGX49400)
