A Fast Surrogate for Modeling Deep-Water Assets with Complex Export Network to Enable Exploration Portfolio Management

Yan-Qing Wang, Daegil Yang, B. Guyaguler, Rong-Lei Zhang, Chang-Dong Yang, Jian-Ping Chen

Abstract

Deep-water exploration portfolio planning often involves hundreds of prospects, numerous tieback platforms, and complex production and export networks. Conventional reservoir–surface network coupled models are computationally intensive and slow at both subsurface simulation and surface network hydraulics calculation, making it impractical for large-scale portfolio optimization. This paper presents a computationally efficient surrogate workflow that repurposes a reservoir simulator to model subsurface production and export-network constraints within a single framework. Prospects and existing production forecasts are represented using decline-curve databases, while platform and flow-path capacities are enforced through simulator hierarchy and field-management logic. Neglecting reservoir status modeling and network pressure-drop calculation, the workflow focuses on production allocation and capacity constraints, eliminating the need for a dedicated surface-network simulator while preserving the key decision drivers for portfolio management. An enhanced allocation capability is introduced to model more complex networks with multiple routing options and predefined flow-path priorities. The workflow has been applied to portfolios containing 50 prospects and 11 platforms. Results demonstrate that key network constraints and production-allocation logics can be accurately represented for both single-routing export network and multi-routing export network, while achieving simulation runtimes of less than five minutes per case, enabling the evaluation of thousands of portfolio optimization scenarios. Although the examples in this paper focus on a single oil phase for workflow demonstration, the underlying formulation, decline-curve representation, capacity constraints, and allocation logic are phase-general and can be extended to multiphase production scenarios, including oil, gas, and water phases.

Source: semanticscholar

Read the AI summary, key takeaways and discussion on WOBR Quant Research.