The invention discloses an elevator life prediction method based on LSTM-ED, and relates to the technical field of elevator life prediction. Comprising the steps that safety indexes, historical data and actual measurement data of an elevator are obtained, the historical data and the actual measurement data of the elevator are split according to the safety indexes, historical time series data and actual measurement time series data are constructed, the historical time series data are subjected to noise reduction and divided into a plurality of subsequences to form a training set and a test set, and the training set and the test set are obtained. And inputting the training set into the LSTM-ED network for training, inputting the training set into the test set for verification to obtain a verified model, inputting the actually measured time sequence data into the verified model, inputting the obtained data into a full-connection network to predict the residual life, and determining the safety level of the elevator according to the residual life. According to the invention, the characteristic error can be minimized, and the elevator detection precision can be improved.

    本发明公开了一种基于LSTM‑ED的电梯寿命预测方法,涉及电梯寿命预测技术领域。包括获取电梯的安全指标、历史数据和实际测量数据,分别将电梯历史数据和实测数据按照安全指标拆分,构建历史时间序列数据和实测时间序列数据,将历史时间序列数据降噪划分为多个子序列形成训练集和测试集,将训练集输入LSTM‑ED网络中进行训练后输入测试集验证,得到验证后的模型,将实测时间序列数据输入验证后的模型中,得到的数据输入全连接网络预测剩余寿命,根据剩余寿命确定电梯安全等级。本发明有助于最小化特征误差,提高电梯检测精度。


    Access

    Download


    Export, share and cite



    Title :

    Elevator life prediction method based on LSTM-ED


    Additional title:

    一种基于LSTM-ED的电梯寿命预测方法


    Contributors:
    ZHENG LIANGTIAN (author) / LIANG ZHIHUANG (author) / LI GUANGLING (author) / YE QIN (author) / CHEN JING (author) / XIE LIUHUI (author) / HE RUOQUAN (author)

    Publication date :

    2025-03-28


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    B66B Aufzüge , ELEVATORS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    LSTM LSTM-based future threat prediction method and apparatus

    PARK YOUNG TACK / JEON MYUNG JOONG / KIM MIN SUNG et al. | European Patent Office | 2021

    Free access

    IGBT Life Prediction Based on CNN1D-LSTM Hybrid Model

    Liu, Qiuli / Tong, Qingbin / Wang, Lei et al. | TIBKAT | 2022


    IGBT Life Prediction Based on CNN1D-LSTM Hybrid Model

    Liu, Qiuli / Tong, Qingbin / Wang, Lei et al. | British Library Conference Proceedings | 2022


    Freight elevator service life prediction device and method

    ZHAO JIANPU / CAO QIAN / LI WENHUI et al. | European Patent Office | 2023

    Free access

    IGBT Life Prediction Based on CNN1D-LSTM Hybrid Model

    Liu, Qiuli / Tong, Qingbin / Wang, Lei et al. | Springer Verlag | 2022