Risk modeling of gas pipeline availability

Abstract

Presently, there is worldwide consideration of Hydrogen pipelines as sustainable energy carriers as well as Carbon Dioxide pipelines for use in achieving net-zero goals through carbon capture and sequestration. For the purposes of planning expansions or new pipelines, typical design criteria like compressor maps, driver loads, etc., are used for simulations of pipeline capacity; however, it is often assumed that the compressor drivers work 100% of the time. In real life, each driver will have an associated availability metric. The availability metric, which parameterizes unit risk of failure, must be accounted for in simulations and pipeline planning to give an accurate view of pipeline capacity. Complicating the analysis is the fact that not all units have equal effect on the pipeline capacity.

In this paper we formalize the framework for including unit availability into pipeline capacity planning and define Pipeline Availability. Availability estimates from industry reports as well as anonymized data from Solar Turbines' global fleet are provided and compared. A novel application of probability theory is used to calculate pipeline availability, and a comparison is made with previous methods that relied on Monte Carlo simulations. Three example applications are presented to show how the novel method is more accurate and much less time consuming than Monte Carlo simulation. Our application of pipeline availability calculations make it easier and more time efficient to consider wide variations of design during the planning and risk evaluation of new Hydrogen or Carbon Dioxide pipelines or expansions of existing Natural Gas pipelines.