In this paper we present our results, challenges and learnings, over a two-year period wherein robust multi-objective optimization was applied at the Mariner asset which is being currently developed. Many different problems were solved with different objectives. These problems were formulated based on the phases of planning and development at the asset. The optimization problems include drilling order and well trajectory optimization as the main objectives with reduction in water cut and reduction of gas production to minimize flaring as secondary objectives. We use the efficient stochastic gradient technique, StoSAG, to achieve optimization incorporating geological and petrophysical uncertainty. For some problems computational limitations introduced challenges while for other problems operational constraints introduced challenges for the optimization. Depending on the problems significant increases between 5% and 20% in the expected value of the objective function were achieved. For the multi-objective optimization cases we show that non-trivial optimal strategies are obtained which significantly reduce (40% decrease) gas production with minimal loss (less than 1%) in the economic objective. Our results illustrate the importance of flexible optimizations workflows to achieve results of significant practical value at different stages of the planning and development cycle at an operational asset.
One important challenge working in heavy oil field development is to present a set of production profiles representing low, base and high production without being overly optimistic or pessimistic. The paper give examples on how various methods of assessing the uncertainty in the relative permeability curve shapes changes the view on project uncertainty. Heavy oil laboratory relative permeability experiments face many well-known challenges. The core data is usually unconsolidated with very high permeability. This combined with high viscosity usually leads to unfavourable laboratory conditions like in example capillary end effects and viscous fingering. Typical core issues are fractures, laminations, loose sand, issues with use of multiple core plugs and very low pressure differential at initiation of the experiment. In addition there is always a discussion about reservoir wettability and core history including impact of drilling operations, drilling mud and core depressurization. The total effect is that the laboratory data is often seen as highly unreliable. The challenges are usually solved by using either light oil or by using steady state together with unsteady state experiments. Weaknesses and strength of the steady state method is addressed. Discussion of the alternative methods that is available is given with examples. Can light oil really be used instead of heavy oil given the large compositional differences? What could the impact of asphaltene content be? Examples are given of some typically laboratory effects. A brief discussion is given regarding possible reasons for differences between light and heavy oil behaviour. A benchmark of relative permeability experience from 50 heavy oil fields is presented. The benchmarking methodology is given and explained. Strength and weaknesses of the method is discussed. Actual field data is shown. Examples of relative permeability data used in simulation models are shown. Impact of grid size and properties on relative permeability is shown. Use of rock curves and alternative curve fit methods is presented with weakness and strengths.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.