The invention discloses a multi-scale convolutional capsule network elevator fault diagnosis method based on data enhancement and an attention mechanism. The method comprises the following steps: 1, acquiring an original vibration signal of the operation acceleration of an elevator car; 2, obtaining an effective vibration signal; 3, optimizing a DCGAN network model, and introducing a spectrum normalization and space-time attention mechanism module into a discriminator and a generator to form an improved data enhancement overall model; 4, inputting the two-dimensional time-frequency feature image into the data enhancement overall model to obtain a generated new feature image; 5, constructing a multi-scale convolutional capsule network; 6, constructing an improved channel attention mechanism (CBAM) network module, and fusing the improved CBAM network module into a multi-scale convolutional capsule network; and 7, sending the enhanced feature map into a multi-scale convolutional capsule network fused with an attention mechanism to carry out elevator fault diagnosis. According to the invention, the feature information is enriched, and the response and learning ability of important features is improved.

    一种基于数据增强和注意力机制的多尺度卷积胶囊网络电梯故障诊断方法,包括以下步骤:第一步:采集电梯轿厢运行加速度的原始振动信号;第二步:获取有效的振动信号;第三步:优化DCGAN网络模型,将谱归一化和时空注意力机制模块引入鉴别器和生成器,构成改进的数据增强整体模型;第四步:将二维时频特征图像输入数据增强整体模型,得到生成的新特征图像;第五步:构建多尺度卷积胶囊网络;第六步:构建改进的通道注意力机制CBAM网络模块,将其融合到多尺度卷积胶囊网络中;第七步:将增强后的特征图送入融合注意力机制的多尺度卷积胶囊网络中进行电梯故障诊断。本发明丰富了特征信息,提高了重要特征的响应和学习能力。


    Access

    Download


    Export, share and cite



    Title :

    Multi-scale convolutional capsule network elevator fault diagnosis method based on data enhancement and attention mechanism


    Additional title:

    基于数据增强和注意力机制的多尺度卷积胶囊网络电梯故障诊断方法


    Contributors:
    LU JIAWEI (author) / ZHANG WEICHAO (author) / WANG QIBING (author) / XIAO GANG (author) / LU CHAO (author) / LI REN (author) / HE YUCHEN (author) / CHEN HANYUAN (author)

    Publication date :

    2024-08-09


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06V / B66B Aufzüge , ELEVATORS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Elevator fault diagnosis method based on convolutional neural network

    ZHANG YUEHONG / YUAN ZHAOCHENG / CHEN FANG et al. | European Patent Office | 2025

    Free access

    Elevator fault diagnosis method based on one-dimensional convolutional neural network and meta learning

    DONG TIANWEI / ZANG CHUANZHI / ZHAO BAOJUN et al. | European Patent Office | 2023

    Free access

    Elevator fault diagnosis device and elevator fault diagnosis method

    TAKAGAMI RYOMA | European Patent Office | 2022

    Free access

    Elevator fault diagnosis method based on big data

    NIU YONGHONG / WU WEI / ZHAO WEI et al. | European Patent Office | 2023

    Free access

    Elevator fault diagnosis method based on graph neural network

    HE YUCHEN / HU WENHAO / WANG QIBING et al. | European Patent Office | 2023

    Free access