YANG Na, LI Tian-hao, ZHAO Gong-min. HEALTH ASSESSMENT OF TIBETAN ANCIENT WOOD STRUCTURES BASED ON THE PREDICTED VALUE OF STRAIN MONITORING DATA[J]. Engineering Mechanics, 2023, 40(2): 112-123, 134. DOI: 10.6052/j.issn.1000-4750.2021.08.0636
Citation: YANG Na, LI Tian-hao, ZHAO Gong-min. HEALTH ASSESSMENT OF TIBETAN ANCIENT WOOD STRUCTURES BASED ON THE PREDICTED VALUE OF STRAIN MONITORING DATA[J]. Engineering Mechanics, 2023, 40(2): 112-123, 134. DOI: 10.6052/j.issn.1000-4750.2021.08.0636

HEALTH ASSESSMENT OF TIBETAN ANCIENT WOOD STRUCTURES BASED ON THE PREDICTED VALUE OF STRAIN MONITORING DATA

  • In order to increase the safety redundancy of the Tibetan ancient wood structures (TAWS), this paper proposes a prediction method to predict their health status based on short-term strain monitoring data. The corridor structure is taken as the research object, with the Sobel operator and the convolutional neural network method being used to deal with various data anomalies from sensors. The strain monitoring data components are determined by polynomial regression and power spectral density. The prophet algorithm is used to predict short-term temperature induced strain, and the Gumbel extreme value theory is used to predict human induced strain under different return periods. The superposition results of the strain monitoring data prediction values under different return periods are used to determine the structural positions that desire more attention and repair, so that the preventive protection requirements and minimum disturbance requirements can be fulfilled.
  • loading

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return