Abstract
New IoT standalone leak detection systems can be rapidly deployed to cover potential leaks, however covering an entire pipeline is cost prohibitive so a method for selecting the locations for deployment is required.
This paper examines how data science & machine learning software combined with environmental assessments can lead to deployment of targeted leak monitoring devices that employ nanotechnolgy based coatings to optimize a pipeline repair program.
When a risk is noted on a pipeline, currently there is little recourse between doing an expensive direct assessment on that location; or letting that location remain unmonitored until resources exist to examine it. New IoT targeted leak detection technology allows an economical middle ground where many risk locations can be monitored for minute traces of signs of a leak, all at a fraction of the price of a direct assessment. With such timely alerting of a leak, the consequences at that site can be greatly reduced. These leak detectors only work in a very localized area and depend on areas of concern being accurately identified. It is modern, advanced ILI alignment and integrity analysis software that provides this key risk placement data.
This leak detection technology uses smart coatings to detect Hydrocarbons, it runs 24/7 and uses alarm-based algorithms to alert to the presence of leaks. It can be rapidly installed in targeted locations that are identified by algorithms as being needed to be repaired and then moved to new locations as those identified defects are repaired.
This paper describes how a fleet of leased mobile leak detection systems were deployed over several months along a pipeline which covered several US States. The units were deployed in High-risk areas that had been identified as needing repair. At the completion of the repair program most of the leased leak detection systems were returned. Feedback from the operator will be included in the presentation.
Available after the event



