SUMMARY

Defect detection and localization are key to preventing environmentally damaging wellbore leakages in both geothermal and oil/gas applications. In this work, a multi-step, machine learning-based approach is utilized to localize two orientations of thermal defects within a wellbore model. This approach includes a COMSOL heat transfer simulation to generate base data, a neural network to classify defect orientations, and a localization algorithm to synthesize sensor estimations into a predicted location. A physical test bed was created to verify the approach using experimental data. The test bed is a small-scale wellbore model with horizontal and vertical copper tubing inclusions. Consistent thermal defects were created by pumping heated water through the inclusions, which were detected by 30 thermocouples adhered to the test bed interior. The localization approach was then able to predict the locations of the copper inclusions.

Bluejeans meeting link: https://bluejeans.com/627511444