基于BO-SDAE多源信号的风电机组轴承故障诊断

Fault Detection of Wind Turbine Bearing Based on BO-SDAE Multi-source Signal

风电机组传感器受环境干扰,导致采集的信号存在差异,从而对风电机轴承的故障诊断结果产生影响。为提高故障诊断的可靠性,提出一种多源信号故障诊断方法。提取轴承振动信号、噪声信号、温度信号的时域、频域特征作为故障特征,利用经贝叶斯优化算法优化隐藏层节点结构的堆叠降噪自编码器对故障特征进行融合,采用Softmax对融合的故障特征进行分类。实验表明:该方法的故障准确率比单一信号进行故障诊断的方法更高,并且混合转速作为实验数据的情况下仍保持较高的准确率。

Due to the discrepancy within signals from sensors of wind turbines caused by environmental interference,the fault detection results of wind turbine bearing will be affected and the multi-source signal fault diagnosis method is proposed to improve the reliability of fault detection.The time-domain and frequency-domain features of bearing vibration signals,noise signals and temperature signals are used for feature extraction,and then the features are transmitted to the stacked denoising autoencoders,which are optimized the hidden layer node structure by the Bayesian optimization algorithm to achieve multi-source signal feature fusion.Softmax function is used for classification.Experiments show that the accuracy of this method is higher than that of the single signal fault diagnosis method,and still maintains a high accuracy rate with mixed speed as experimental data.