基于盒维数和参数筛选的风力机轴承故障诊断方法研究

Research on fault diagnosis method of wind turbine bearing based on box dimension and parameter selection

风力发电机作为一种清洁能源被广泛地应用,轴承作为风电机组的核心零部件故障诊断极为重要,针对风力机轴承故障诊断的问题,提出了一种基于盒维数计算和参数筛选的故障诊断方法,并通过支持向量机进行训练和分类。首先,通过盒维数算法计算盒维数和其它特征参数,然后筛选正确参数并分类得到各组参数的准确度,最后确定不同故障形式下样本的特征,最后筛选出准确度较高的特征参数,避免了复杂的信号预处理以及人工进行信号提取的过程,可以高效解决风力机轴承故障诊断的准确率低、时效性差等问题。

As a kind of clean energy,wind turbine is widely used,and fault diagnosis of bearing as the core component for wind turbine is extremely important.For the problem of fault diagnosis of wind turbine bearing,a fault diagnosis method based on box dimension calculation and parameter screening is proposed,which is trained and classified by support vector machine.First,calculate the box dimension and other feature parameters through the box dimension algorithm,then screen the correct parameters and classify them to obtain the accuracy of each group of parameters,finally determine the characteristics of samples under different fault forms and the feature parameters with high accuracy,avoiding the complex signal preprocessing and manual signal extraction process,which can effectively solve the problems of low accuracy and poor timeliness of wind turbine bearing fault diagnosis.