Elevator guide shoe fault diagnosis method

You are here: Homepage > Search

Elevator guide shoe fault diagnosis method

Free access

LI KUN / ZHU XIAOLING

The invention relates to an elevator guide shoe fault diagnosis method, and belongs to the technical field of special equipment fault diagnosis. The elevator guide shoe fault diagnosis method is a diagnosis method based on Gabor wavelet transform and a multi-core support vector machine, so that more accurate comprehensive diagnosis can be carried out on elevator guide shoe faults. Firstly, a vibration signal of an equipment body is collected through an acceleration sensor, and an intrinsic mode function component is obtained through empirical mode decomposition; then, a Gabor filter is adopted to carry out filtering and denoising on the low-frequency component, and feature enhancement of data extracted at a low frequency is achieved; and finally, linear addition is carried out on the local and global kernel functions by adopting a weight mode to form a multi-core support vector machine to classify the data. An experimental result verifies the effectiveness of the method. Compared with a fault diagnosis method based on a wavelet transform least square support vector machine, the fault diagnosis accuracy of the method is improved by about 5%.

本发明涉及一种电梯导靴故障诊断方法，属于特种设备故障诊断技术领域。为了对电梯导靴故障进行更加准确的综合诊断，提出了一种基于Gabor小波变换和多核支持向量机的诊断方法。首先，通过加速度传感器采集设备主体的振动信号，并利用经验模态分解得到固有模态函数分量。然后，采用Gabor滤波器对低频分量进行滤波去噪，实现低频率上提取数据的特征增强。最后，采用权重的方式将局部和全局的核函数进行线性相加，组成了多核支持向量机对数据进行分类。实验结果验证了提出方法的有效性。相比基于小波变换‑最小二乘支持向量机的故障诊断方法，提出方法的故障诊断准确率提高了约5％。