University of Twente Student Theses

Login

Quantifying thermal imaging data for piping detection : analyzing the flow rate and the size of the temperature spots in thermal images of seepage at dykes

Klaarbergen, I.C. van (2022) Quantifying thermal imaging data for piping detection : analyzing the flow rate and the size of the temperature spots in thermal images of seepage at dykes.

[img] PDF
10MB
Abstract:The failure mechanism piping is a significant problem for the safety of river and sea dykes located in delta areas. This is the case in large parts of the Netherlands. Piping is a process that can reach an advanced stage before any sign is remarkably visible. Therefore, it is important to monitor the dykes to have insight in the current conditions with respect to piping. The measurement of temperature is proven to be an effective tool for the detection of seepage spots which are possible piping locations in dykes. With the use of a camera with infra-red sensors, the surface temperature of structures can be measured, resulting in a thermal image. With thermal imaging, potential problematic seepage locations can be mapped very fast and efficiently since seepage often has a different temperature than its surrounding. With human inspection they are sometimes missed due to a coverage of grass or water plants. The application of thermal imaging for management and inspection of flood defenses is still relatively limited. The technique is used to detect locations of seepage spots in the field but more than the location is not derived yet. The quantification of thermal imaging data for piping detection is a subject that needs to be investigated. In this study, the sizes of the seepage spots in the thermal images and the relation between these spots and the flow rate through the sand boils is quantified. The aim is to determine to what extent thermal imaging can be used to make a flow rate prediction of a seepage spot. The case study which is used for this research is the full-scale piping experiment in a controlled field environment at the Living Lab Hedwige-ProsperPolder (LLHPP) which was performed in September of 2021. To determine to what extent thermal imaging can be used to make a flow rate prediction of a seepage spot, three main steps are taken. Firstly, a data analysis method is developed to quantify the sizes of the areas of the found seepage spots in the thermal images. Secondly, a correlation analysis is performed to find out if there is a relation between the areas of the seepage spots in the thermal images and the flow rate of the seepage through the sand boils. Lastly, a regression analysis is performed to develop a practical tool to enable flow rate prediction based on the size of the seepage spot in the thermal images. For a large part of the thermal imaging data it was possible to quantify the seepage spots. The best circumstances to quantify the seepage spots in the thermal images are found when the influence of the sun on the surface temperature of the ditch is not significant. So, during night or when the weather is not too sunny. The found correlation showed promising results. The correlating part of the data had little influence of sunlight, there were no people in the camera view and one sand boil was mainly active. This part showed a strong correlation between the flow rate and the size of the seepage spot in the thermal image which can be used for a regression analysis. The regression analysis resulted in a practical tool which enables flow rate prediction based on the size of a seepage spot in a thermal image. The regression analysis showed that the flow rate increases slower at areas which are bigger, and faster at areas which are smaller. The results of this study can be implemented in practice as an additional tool to complement the inspection of dykes for piping. It makes human observations more accurate, supports the interpretation of what is seen and can find critical seepage locations which human observations can miss upon. In conclusion, the results that are found in this study show that thermal imaging can be used to make a prediction of the flow rate of a seepage spot under certain circumstances. It is a very promising method to be used as a complementary tool to enhance the quality of the dyke assessment.
Item Type:Essay (Master)
Faculty:ET: Engineering Technology
Programme:Civil Engineering and Management MSc (60026)
Link to this item:https://purl.utwente.nl/essays/92012
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page