Determining Flushing Efficacy on Pipelines Using Ultrasonic Sensors

Abstract Flushing was being utilized as an internal corrosion mitigation in relief lines of crude pipeline facilities; however, the efficacy of flushing for removing settled bottoms, sediment, and water in the relief piping was unknown. A project was developed to utilize permanently installed ultrasonic sensors to characterize what occurs during a flush event and whether that event was effective at removing deposits and reducing corrosion rates. Ultrasonics are normally utilized for measuring wall thickness in steel pipe, but sound waves travel beyond the back wall of the steel. Differences in the ultrasonic wave forms are observable in pipes filled with varying amounts of water, oil, and sediment. Waveforms of different conditions of water, oil, and sludge were gathered in the lab and used to train an artificial intelligence (AI) model which was utilized to interpret waveforms of the permanently installed sensors in the field before, during, and after flush events.