An elevator vibration fault diagnosis method based on car acceleration analysis comprises the steps that firstly, the elevator operation process and typical vibration faults are described, secondly, feature extraction is conducted on elevator car vibration signals from a time domain, a frequency domain and a time domain, and feature vectors of vibration samples are reconstructed on the basis; and finally, combining with a single gating unit recurrent neural network (SGU) and a convolutional neural network (CNN) to construct a CNN-SGU model so as to realize the intelligent diagnosis of the excessive vibration fault and the sudden stop fault of the elevator. According to the method, the classification precision can be improved, and real-time diagnosis of normal operation, excessive vibration faults and sudden stop faults of the elevator is achieved.
一种基于轿厢加速度解析的电梯振动故障诊断方法,首先描述了电梯运行过程及典型振动故障,其次分别从时域、频域及时频域对电梯轿厢振动信号进行特征提取,并在此基础上重构了振动样本的特征向量,最后将与一种单门控单元循环神经网络SGU卷积神经网络CNN组合,构建CNN‑SGU模型实现了对电梯振动过大故障、急停故障的智能诊断。本发明有助于提高分类精度,实现了对电梯正常运行、振动过大故障以及急停故障的实时诊断。
Elevator vibration fault diagnosis method based on car acceleration analysis
一种基于轿厢加速度解析的电梯振动故障诊断方法
2023-12-29
Patent
Electronic Resource
Chinese
Elevator fault diagnosis device and elevator fault diagnosis method
European Patent Office | 2022
|