Automated reservoir management using multi-objective simulation optimization and SA model

作者：Moradi Tanaz; Rasaei Mohammad Reza^*

来源：Journal of Petroleum Science and Engineering, 2017, 150: 91-98.

DOI：10.1016/j.petrol.2016.11.024

• Situation awareness
• Automatic oil production planning
• Multi-objective optimization
• Water flooding
• Pareto-front
• Reservoir management

Technology advancements steer implication of automation in the cognitive functionality of reservoir management, emerging chiefly in optimization and its peripheral concepts. Smart field development has been modeled upon real-time management, consequently, closed-loop management has taken a step further integrating data assimilation and optimization in a near-continuous process. Nevertheless, automation in planning domain threats the concealed concept of Situation Awareness (SA) which is a predominant planning concern in complex dynamic systems.
SA incorporates the perception of environmental elements and events, the comprehension of their meaning, and the projection of their status in future to make conscious decisions. In current practices of smart reservoir management, evaluation of reservoir status and setting-up future goals are temporally accomplished upon interactions of engineers and simulation optimization. However, this traditional effort to preserve SA upon human interference results in descent of automation and suspension of its demonstrated beneficial advantages in real practices.
An effective and practical automated reservoir management should self-sufficiently consider SA in automation mechanism. In this paper, the goal is to present an innovative theoretical model of automated reservoir management, which considers SA as a mindful gadget that pertinently define multi-objective optimization problem, regarding reservoir dynamic situation in line with economic, technical and operational managerial targets.
The proposed model is compatible with previous models and moreover supplements them with a conscious automatic planning potential. Utilizing this model, multi-objective optimization will provide higher level of automation upon definition of temporal objectives and preferences without human interference. This model introduces higher degree of freedom in decision making and generates production/injection profiles flexible with predefined objectives and preferences of the management, which is advocated through a case study of water injection reservoir. Conceivable sets of dynamic objectives in different optimized production/injection profiles illustrate the establish role of SA in automatic reservoir management.

• 出版日期2017-2