A 22-year-old liquid line experienced an unexpected failure due to internal corrosion after adding the production from one well. Investigation included ILI records, operating pressures and temperatures, oil pressure / volume / temperature (PVT) data, possible flow regimes, failure analysis reports, and mitigation practices.
Simulation shows that navigation algorithms can guide autonomous vehicles to inspect poles or towers for corrosion, or to inspect enclosed spaces. The algorithms use sensor feature extraction, PID control, dynamic window obstacle avoidance, and genetic algorithm solutions to the traveling-salesman problem.
It is possible to design, construct, and perform experiments with a modular corrosion loop allowing a detailed investigation of corrosion processes. The instrument described in this paper was aimed at validating the experimental concept via testing a material that is susceptible to erosion-corrosion.