Considering the low and volatile oil and gas prices, cost reductions are an impending challenge in the oil and gas industry. For example, offshore pipelines constitute approximately 12% of total CAPEX and potentially 30% of the total OPEX of the entire offshore field development cost. Workforce shortages and manual engineering processes have a high cost impact due to the excessive time to optimize pipeline design, construction, and operation for a robust and efficient system. If the design phase takes excessive time, it is difficult to optimize cost for CAPEX and OPEX while managing risks, with the added complexity of price uncertainties in the market due to unforeseen geopolitical instability. The objective of this paper is to present an artificial intelligence and automation driven pipeline engineering computation system for performing pipeline design and integrity engineering that can save substantial engineering time with a better and optimized design that can accommodate to crude price uncertainty.
A comprehensive computational framework is utilized for pipeline conceptual/preliminary, FEED and operational integrity analysis. The digital computational framework is equipped with a distributed data intake, a robotics process automation, and a predictive analytics computation system. A self-service 3D digital twin creator is utilized to operate from the front end and receiving intelligence from the analysis. The automated design engineering engine is utilized to perform and create ready reports for conceptual/preliminary and FEED/Detail design of a subsea pipeline system. A predictive analytics computation system is utilized to perform pipeline operational integrity using in-line-inspection data and life prediction.
The computational system was tested for a subsea flowline conceptual, FEED level design (i.e. pipeline thermo-mechanical and on-bottom-roughness FEA analysis) and thermomechanical structural integrity during the operational phase. The design time was substantially lower than the traditional manual pipeline design suggested a saving as much as 80% of the engineering time. The computation engine was also tested to conduct pipeline integrity using machine learning from direct in-line-inspection data. The flexibility and scalability of the computational framework allow incorporation of any organization's existing pipeline data and engineering process.
The significant saving in the engineering time should be able to reduce contingency in hedging due to the price uncertainty. The novel computational framework is made available online offers real-time on-demand analytics for instant intelligence for making decisions for both in design and in operational phases. Reusing of all engineering efforts over the life of the pipeline or for other projects is another possibility offered by the cloud-based computational framework.