Production Optimization: A Moving-Horizon Approach
- Michael Nikolaou (U. of Houston) | A. Stan Cullick (Halliburton Digital and Consulting Solutions) | Luigi Alfonso Saputelli (Halliburton Digital and Consulting Solutions)
- Document ID
- Society of Petroleum Engineers
- Intelligent Energy Conference and Exhibition, 11-13 April, Amsterdam, The Netherlands
- Publication Date
- Document Type
- Conference Paper
- 2006. Society of Petroleum Engineers
- 4.1.2 Separation and Treating, 5.5 Reservoir Simulation, 7.6.6 Artificial Intelligence, 1.7.5 Well Control, 1.6 Drilling Operations, 2.2.2 Perforating, 5.4.1 Waterflooding, 4.3.4 Scale, 5.5.8 History Matching, 6.5.2 Water use, produced water discharge and disposal, 5.2.1 Phase Behavior and PVT Measurements, 2.3 Completion Monitoring Systems/Intelligent Wells, 4.1.5 Processing Equipment
- 7 in the last 30 days
- 675 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 8.50|
|SPE Non-Member Price:||USD 25.00|
Designing fluid injection policies to optimize the production of a hydrocarbon reservoir has attracted considerable interest in recent years. Production policies can emerge from numerical optimization analyses, the solutions for which are most frequently based on optimal control theory. In this paper we argue that (a) a simpler alternative to the optimal control approach may be used, and (b) we present a moving-horizon formulation alternative. We illustrate the proposed approach through several examples.
The increasing availability of real-time downhole measurements and remotely activated valves in an oil-field has made field-wide optimization of operations in real time a distinct possibility. While the term real-time optimization (RTO) is certainly not new and RTO is practiced in elements of drilling or production operations,[2,3] the extent to which RTO is now feasible has increased dramatically. At the same time, the increased scope of RTO of oil-field operations entails significant complexity and creates challenges related to:
Conceptual development: e.g., what is "real time??? What should be optimized? What are associated work flows?
Technological realization: e.g., what hardware and software should be used? When? Where?
Practical implementation: e.g., what is the expected and actual return on investment?
Management: e.g., who is responsible for the development, implementation, operation, and maintenance?
RTO technologies have been advanced, either within the oil and gas industry or in related industries, such as oil refining. While it would certainly be beneficial to further develop technologies for field-wide RTO, it would also be useful to identify existing technologies suitable for the task, streamline such technologies for use in the oil-field, and ensure that such technologies are used prudently and ultimately add value.[5,6] Because elements of field-wide RTO can be manifest in many activities related to production optimization,[7,8] one may be overwhelmed by the multitude of approaches and breadth of scope of field-wide RTO. Putting field-wide RTO in a concrete framework offers clear development and implementation benefits, in that it can catalyze progress by suggesting the path to long-term benefits that might not be immediately obvious from incremental improvements from individual projects.
Building on previous work that established the multi-scale nature of RTO and focused on decision making at the time-scale of days to weeks, we are concentrating in this work on decision making at coarser time-scales, e.g., months, with application to optimizing the production of a hydrocarbon reservoir by proper injection of fluids. Capitalizing on significant new capabilities for bottom-hole measurements and remote valve manipulation, a number of authors have convincingly pointed out that sizeable economic benefits can result by proper selection of fluid injection policies (for a thorough recent review see Brouwer's Ph.D. Thesis). Such policies can be designed by solving a numerical optimization problem, usually for the net present value (NPV) of a project or for total oil recovery. The numerical solution of such an optimization problem has attracted considerable attention. The approach most frequently taken is based on optimal control theory. In this work we provide a critical assessment of recent work in this area, identify complexity issues, and suggest approaches toward complexity reduction based on the concept of moving-horizon optimization. In particular, we argue that (a) simpler alternatives to the optimal control approach may be used and (b) the formulation of a selected optimization problem may significantly affect how efficiently that problem can be solved. We illustrate the proposed approach through a number of examples.
|File Size||1 MB||Number of Pages||13|