基于NIC-DWT-WOASVM的齿轮箱混合故障诊断

由于齿轮箱中振动信号的复杂性和非平稳性,致使齿轮箱混合故障诊断工作具有一定难度。针对这一问题提出基于NIC-DWT-WOASVM的齿轮箱混合故障诊断方法。首先通过窄带干扰消除(Narrow Band Interference Canceller, NIC)滤除原始信号中齿轮啮合和转轴等窄带干扰信号,接着对信号进行离散小波变换(Discrete Wavelet Transform, DWT),重构小波系数得到小波分量,提取分量的方差作为特征参数构成特征矩阵样本。针对传统优化支持向量机收敛速度慢及容易局部最优等问题,提出鲸鱼算法优化的支持向量机(Whale Optimization Algorithm Support Vector Machine, WOASVM),运用训练样本对WOASVM进行训练得到优化分类模型,将测试样本输入到优化模型中得到诊断结果。为验证方法的有效性,开展了变工况下齿轮箱混合故障实验,通过实验分析及与其他方法的比较,证明方法对于齿轮箱混合故障诊断是有效的。

Compound fault diagnosis for gearbox based on NIC-DWT-WOASVM

The compound fault diagnosis of gearbox is challenging because of its complexity and non-stationarity of the vibration signal. In this work, a novel hybrid method based on narrow band interference canceller (NIC), discrete wavelet transform (DWT) and support vector machine optimized by whale optimization algorithm (WOASVM) is presented for compound fault diagnosis of gearbox. Firstly, the raw signal is processed by NIC to filter the deterministic signal which interfere the fault signal. Then the signal is processed by discrete wavelet transform, the wavelet coefficients are reconstructed and the variances of the wavelet components are calculated as the characteristic parameters. Aiming at the problems of slow convergence speed and easy local optimization of traditional optimized support vector machines, the WOASVM is proposed for fault pattern recognition. The training samples is used to train WOASVM and obtain the optimized classification model, and input the test samples into the optimization model to get the diagnosis results. In order to verify the effectiveness of the proposed method, a compound fault experiment of gearbox under variable conditions is carried out. Via experimental analysis and comparison with other methods, it is proved that the method is effective for the compound fault diagnosis of gearbox.