基于SVD转子不平衡不对中故障的轴心轨迹识别

Axis Trajectory Identification Based on SVD Rotor Unbalance Misalignment Fault

针对转子轴心轨迹存在噪声、共振等外部干扰问题,提出了一种基于奇异值分解(SVD)对转子信号进行降噪提纯的方法。通过采集转子轴向信号X和径向信号Y构造Hankel矩阵,对矩阵进行SVD分解,根据奇异值差分谱峰值选取有效奇异值来重构信号,将重构后信号合成轴心轨迹完成降噪提纯过程。经仿真分析,SVD能有效对信号进行降噪提纯。在转子实验中,将SVD用于转子故障轴心轨迹提纯。实验结果表明,经SVD降噪后的轴心轨迹清晰可见,转子不平衡的轴心轨迹为椭圆形,不对中的轴心轨迹为香蕉形,与理论相符。因此,SVD算法能有效提纯轴心轨迹并成功识别转子故障。

Aiming at the external interference problems such as noise and resonance in the rotor axis trajectory,a noise reduction and purification method based on SVD(Singular Value Decomposition)is proposed. By acquiring the rotor axial signal X and radial signal Y, the Hankel matrix is constructed, the matrix is SVD decomposed, the effective singular value is selected to reconstruct the signal according to the peak of the singular value difference spectrum, and the reconstructed signal is synthesized into an axis trajectory to complete the noise reduction and purification process. After simulation analysis, SVD can effectively purify the signal for noise reduction. In rotor experiments, SVD is used for rotor fault axial trajectory purification.Experimental results show that the axis trajectory after SVD noise reduction is clearly visible, the axis trajectory of rotor imbalance is elliptical, and the misaligned axis trajectory is banana-shaped, which is consistent with the theory. Therefore, the SVD algorithm can effectively purify the axis trajectory and successfully identify rotor faults.