WANG Mingming, WANG Sha, XING Hui, SUN Xiaoyun, LU Lin. The application of stacking auto-encoder in the identification of bolt anchoring defects[J]. CHINA MINING MAGAZINE, 2020, 29(7): 81-85. DOI: 10.12075/j.issn.1004-4051.2020.07.023
    Citation: WANG Mingming, WANG Sha, XING Hui, SUN Xiaoyun, LU Lin. The application of stacking auto-encoder in the identification of bolt anchoring defects[J]. CHINA MINING MAGAZINE, 2020, 29(7): 81-85. DOI: 10.12075/j.issn.1004-4051.2020.07.023

    The application of stacking auto-encoder in the identification of bolt anchoring defects

    • In order to solve the problem that the traditional feature extraction method relies on human experience and can’t mine the deep features of data and reduce the accuracy of anchor bolt defect identification, this paper proposes a bolt anchoring defect identification algorithm based on automatic layer selection stacking auto-encoder feature extraction.The algorithm first optimizes the reconstruction error by using Adam optimization algorithm, and automatically determines the depth and parameters of the stacking auto-encoder network, so as to effectively improve the sensitivity of extracted features to defects.Then, the Softmax multi-classifier is used to identify the anchoring defects of the extracted feature signals.Finally, the algorithm is verified by numerical simulation and physical simulation.The results show that the feature extraction method based on the automatic layer selection stacking auto-encoder can effectively extract the bolt anchoring defect features, making the average recognition rate of numerical simulation and physical simulation defects reach over 97%.
    • loading

    Catalog

      Turn off MathJax
      Article Contents

      /

      DownLoad:  Full-Size Img  PowerPoint
      Return
      Return