The invention provides an elevator fault diagnosis method based on a one-dimensional convolutional neural network and meta-learning, which comprises the following steps: firstly, acquiring an acceleration signal generated in the movement process of an elevator car, then carrying out preprocessing and wavelet soft threshold noise reduction on the acceleration signal, extracting the contour feature of the data after noise reduction, fixing the data dimension by using a resampling technology, and finally obtaining the fault diagnosis result of the elevator car. The method comprises the steps of generating a complete data set, dividing the complete data set according to an N-way K-shot principle to enable the complete data set to meet the requirements of a meta-learning strategy, sending the divided data into a one-dimensional convolutional neural network in batches, and completing the training and testing of a model by adopting the meta-learning strategy. And an elevator fault diagnosis model with strong generalization ability and high precision can be obtained by using a small number of samples.
本发明提供了基于一维卷积神经网络和元学习的电梯故障诊断方法,先获取电梯轿厢在运动过程中产生的加速度信号,然后对加速度信号进行预处理和小波软阈值降噪,提取降噪后数据的轮廓特征,使用重采样技术固定数据维数,生成完整数据集,按照N‑way K‑shot原则划分完整数据集使其符合元学习策略的要求,将划分好的数据分批次送入一维卷积神经网络中采用元学习策略完成模型的训练与测试,该方法能够有效降低电梯运行特征的提取难度,使用少量样本就可获得具有很强泛化能力、较高精度的电梯故障诊断模型。
Elevator fault diagnosis method based on one-dimensional convolutional neural network and meta learning
基于一维卷积神经网络和元学习的电梯故障诊断方法
2023-01-06
Patent
Electronic Resource
Chinese
Elevator fault diagnosis method based on convolutional neural network
European Patent Office | 2025
|Elevator fault diagnosis method based on graph neural network
European Patent Office | 2023
|Elevator fault diagnosis device and elevator fault diagnosis method
European Patent Office | 2022
|Elevator fault diagnosis system using recurrent neural network model
European Patent Office | 2023
|European Patent Office | 2024
|