Development of prediction and detection technology of spalling for RC structures using 3D point cloud data and deep learning

2023 Fiscal Year Final Research Report

• Project/Area Number: 22K20454
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0303:Civil engineering, social systems engineering, safety engineering, disaster prevention engineering, and related fields
• Research Institution: Anan National College of Technology
• Principal Investigator: Kadono Takuma 阿南工業高等専門学校, 創造技術工学科, 講師 (80963264)
• Project Period (FY): 2022-08-31 – 2024-03-31
• Keywords: 鋼材腐食 / ひび割れ / 剥離・剥落 / 劣化予測 / 機械学習

Outline of Final Research Achievements

In this study, a machine learning model was constructed to regress the depth of steel corrosion on the amount of deformation that occurs on the concrete surface as steel corrosion progresses, and a basic study was conducted to contribute to the development of a method to detect cracks, spalling and delamination caused by steel corrosion in reinforced concrete structures from three-dimensional point cloud data obtained with laser ranging equipment. A database was constructed to organise the relationship between the depth of steel corrosion and the associated vertical displacement of the concrete surface, based on the results of analyses using a continuum damage model based on fracture mechanics, in which the cover and other parameters were varied. A regression model for the steel corrosion depth was developed using the database as teacher data and with the aid of a neural network.

Free Research Field

コンクリート工学

Academic Significance and Societal Importance of the Research Achievements

鋼材腐食によるかぶりコンクリートの剥離・剥落予測手法として鋼材腐食深さを指標としたモデルが提案されているものの，鋼材腐食速度が構造物ごとに異なることから実務での適用拡大には至っていない．また，鋼材腐食速度を構造物ごとに精緻に把握しようとすると，コンクリートのはつり出しが必要になる場合があることから，非常に労力と時間を要すこととなる．そのため，本研究成果を活用し，構造物の表面の変形性状から鋼材腐食深さを推定することが出来れば，劣化の初期段階において，将来的なかぶりコンクリートのひび割れや剥離・剥落を検知・予測することが可能となるため，学術的意義および社会的意義がある研究であると言える．