The invention provides an elevator time series data anomaly diagnosis method based on deep learning, and belongs to the field of special equipment operation process soft measurement modeling and application, and the method mainly comprises the following steps: determining auxiliary variables needed by elevator soft measurement modeling based on data, and carrying out the normalization processing of the auxiliary variables; segmenting the elevator data into time series data with a fixed length by adopting a sliding window; carrying out trend prediction of the long-short-term memory network based on the attention mechanism on the time series data, and predicting the state value of the next moment; and inputting the state value into a normal range area reconstructed by adopting a variationalencoder, and judging whether the state value is in the normal range area or not, thereby obtaining the abnormal condition of the data. The method not only can improve the accuracy of abnormality diagnosis in the elevator running process, but also can provide a real-time detection effect, and can be effectively applied to the field of fault diagnosis of special inspection equipment.
本发明提供了一种基于深度学习的电梯时序数据的异常诊断方法,属于特种设备运行过程软测量建模和应用领域,主要步骤如下:基于数据确定电梯软测量建模所需的辅助变量并对辅助变量进行归一化处理,采用滑动窗口将电梯数据分割成固定长度的时序数据;对时序数据进行基于注意力机制的长短时记忆网络的趋势预测,预测下一时刻的状态值;将状态值输入到采用变分编码器重构的正常范围区域中,通过判断状态值是否处于正常范围区域中,从而得到该数据的异常情况。该方法不仅能提高电梯运行过程异常诊断的精度,还能够提供实时检测的效果,可有效应用于特检设备的故障诊断领域。
Elevator time series data anomaly diagnosis method based on deep learning
一种基于深度学习的电梯时序数据的异常诊断方法
2021-03-12
Patent
Electronic Resource
Chinese
ELEVATOR ANOMALY DIAGNOSTIC DEVICE AND ELEVATOR ANOMALY DIAGNOSTIC METHOD
European Patent Office | 2023
|European Patent Office | 2023
|