Evaluation of Methods for UDOT Brine Tank Condition Assessment

Details

• Creators:
  Kami, Moein Ramezanpour ; Zhang, Keping ; Chen, Jianli ; Zhu, Xuan
• Corporate Creators:
  University of Utah. Department of Civil and Environmental Engineering
• Corporate Contributors:
  Utah. Dept. of Transportation. Research Division ; United States. Department of Transportation. Federal Highway Administration
• Subject/TRT Terms:
  Brines Computer Vision Condition Assessment Edge Detection Environmental Risk Assessment Groundwater Inspection Machine Learning Storage Tanks Structural Analysis Thermal Imagery Winter Maintenance

  More +
• Publication/ Report Number:
  UT-24.14
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States
• Edition:
  Final Report Aug. 2021 to Dec. 2023
• Corporate Publisher:
  Utah. Dept. of Transportation. Research Division
• Abstract:
  UDOT manages statewide above-ground storage tanks to store brine as deicing chemicals for the winter maintenance program. The polyethylene storage tanks will age under UV light, become brittle with low temperatures, creep, and develop cracks. Structural failures of brine tanks will not only impose safety hazards to surrounding tanks and UDOT personnel but could also contaminate groundwater. While these tanks are routinely inspected to identify signs of aging and damage, current methods may not adequately capture two of the leading causes of tank failure: (a) small defects and wall thinning that can finally result in rupture and leakage, and (b) stress concentration due to inappropriate installation or uneven base. Therefore, it is essential to develop rapid inspection technologies along with guidelines as an implementable solution for UDOT’s routine inspection. The recent advancement of computer vision (CV) and machine learning (ML) has been developed to be capable of automatically extracting temperature information from thermal images and evaluating deflection based on optical images. In this project, we developed accurate and efficient CV and AI algorithms to automatically identify wall thinning and measure tank wall profile/deflection by analyzing images recorded by a dual-spectrum camera. The use of the proposed technologies can potentially enable an objective full-field structural condition inspection on UDOT brine tanks, which is not available from the current practice. However, this research also establishes recommended routine inspection guidelines for UDOT brine tanks using a low-cost thermal camera rather than the dual-spectrum camera due to cost concerns.
• Format:
  PDF
• Funding:
  22-8115
• Collection(s):
  US Transportation Collection
• Main Document Checksum:
  urn:sha-512:d66e90dcd46b0cc32fda3274ceb21da945c8720557107a3de65a2dd8d62b2653eb58d58c2c336486e10601acdef7c913ba41ac71c03213d30c35fcd0483a0267
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/77449/dot_77449_DS1.pdf
• File Type:
  [PDF - 2.49 MB ]