Title: |
Development and Applications of Production Optimization Techniques for Petroleum Fields |
Author: |
Pengju Wang |
Year: |
2003 |
Degree: |
PhD |
Adviser: |
Aziz |
File Size: |
1343K |
View File: |
|
Access Count: |
1231 |
Abstract:
For some petroleum fields, optimization of production operations can be a major factor on increasing production rates and reducing production cost. In this work we investigated the formulations and solution methods for the following optimization problem: determining the optimal production rates, lift gas rates, and well connections for a gathering system with tree-like structures to maximize daily operational objectives subject to multiple flow rate and pressure constraints. While some aspects of this problem have been studied by other investigators, existing methods are either inefficient or they are based on significant simplifications that lead to suboptimal solutions. Hence, it is necessary to develop approaches that can solve the problem efficiently and without unreasonable assumptions.
We first investigated efficient procedures for simulating the production system and computing sensitivity coefficients of production rates relative to system parameters. These procedures have useful applications in simulation, sensitivity analysis, and optimization of a petroleum field.
We then focused on the rate allocation problem, which refers to the determination of the optimal production and lift gas rates subject to multiple constraints. When flow interactions among different wells are not significant, the well performance can be analyzed individually. Consequently, the rate allocation problem can be formulated as a separable programming (SP) problem whose objective and constraint functions are sums of functions of one variable. The SP problem is solved by various linear optimization techniques. This method is very efficient. However, it may lead to bad solutions when the flow interactions among wells are significant. In such cases, the rate allocation problem should be appropriately formulated so that simulations capable of capturing such flow interactions can be conducted in the optimization process. Several formulations were investigated. The suggested formulation is able to handle well shut-down, avoid some numerical difficulties, and is computationally efficient. Once formulated, the optimization problem was solved by a sequential quadratic programming (SQP) algorithm.
A two-level programming approach is developed to optimize the production rates, lift gas rates, and well connections simultaneously. In this approach, the entire optimization problem is solved in two levels. The upper level masters the overall solution process and explicitly optimizes the well connections. For each new set of well connections explored in the upper level, a lower level problem is formed to determine the optimal set of production and lift gas rates. The lower level problem is a rate allocation problem that can be solved by various optimization methods developed in this study. The upper level problem is solved by a newly developed heuristic method. The method is both efficient and robust, as verified by a genetic algorithm.
Production engineers often strive to achieve multiple conflicting goals when operating a field. Several existing multiobjective optimization methods are used to address this problem. Through an example, this study demonstrated that multiobjective optimization methods can help decision makers identify the best trade-offs. The optimization tools were coupled with models for multiphase fluid flow in reservoirs and surface pipeline networks through a commercial reservoir simulator. The coupled procedure was applied to short-term production optimization in the Prudhoe Bay field of the North-Slope of Alaska, and to long-term reservoir development studies for two fields in the Gulf of Mexico. Results demonstrated the effectiveness of the approach.
Press the Back button in your browser.
Copyright 2003, Pengju Wang: Please note that the reports and theses are copyright to their original authors. Authors have given written permission for their work to be made available here. Readers who download reports from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the author, Pengju Wang.
Accessed by: ec2-3-231-222-84.compute-1.amazonaws.com (3.231.222.84)