摘要
针对光伏并网逆变器的故障检测与智能在线诊断问题,提出了一种C3C3逆变器故障特征提取方法。该方法以逆变器输出三相电流的小波分析结果作为判断依据,将故障信号的近似分量和细节分量相结合作为故障特征向量,利用神经网络的分类功能,实现对光伏并网逆变器的故障诊断。该方法无需电压信号和人工参与,硬件实现简单,且抗干扰性强。仿真实验结果证明了该方法的有效性。
Aiming to the fault detecting of photovoltaic grid-connected inverter and its intelligent online diagnosis problem, it proposed a C3C3 inverter fault feature extraction method, which used a three-phase inverter output current as a result of judgments based on wavelet analysis, and combines the approximate component and detail component of failure signals as failure feature vector; then used the classification of neural network to complete the fault diagnosis of photovoltaic grid inverters. This approach will not require manual intervention and the voltage signal, which has simple hardware implementation and strong anti-interference. The effectiveness of the proposed method has been proved through the simulation experiment.
出处
《大功率变流技术》
2014年第5期12-16,共5页
HIGH POWER CONVERTER TECHNOLOGY
关键词
并网逆变器
故障特征
故障诊断
小波分析
神经网络
光伏发电
grid-connected inverter
fault characteristics
fault diagnosis
wavelet analysis
neural network
PV （photovoltaic）
