Sharif Digital Repository

In recent years, due to the development of smart wells, optimization of waterflooding by injection/production rate control has receivedsome interests. We can postpone breakthrough time and increase the sweep efficiency by using inflow control valves (ICV). Due to complexity of reservoirs, existence of constraints and numerous influencing parameters, we need a robust and suitable optimization approach to overcome such problems. In this thesis, model predictive control (MPC) is chosen to be our optimization approach. MPC is suitable for constrained multi variable functions. Genetic algorithm is was chosenas optimizer. Eclipse reservoir simulator was used for reservoir simulation. Eclipse input... 

Intelligent control or management of well or reservoir is a method which is used in order to optimize the production for increasing recovery factor so that the minimum damage or reduction of pressure occurs during production lifetime. Azar oilfield has 17 wells and the production from this field isn't started. Thereupon designing a suitable well completion model and a control algorithm to optimize the production for the first two years of production is necessary and indispensable. For this purpose, by collecting basic information and their analysis (for example pressure, temperature, production rate, the type and amount of produced fluids), designing of downhole installations and modeling... 

Optimizing and improving the design of wells will have significant impact on any recovery method. As for horizontal wells, despite the fact that horizontal wells improve reservoir access, they suffer from significant pressure loss as well as limited flow control locations and flow control and management. In addition, reservoir quality varies along the length of the well and in the reservoir. This will add further complexity and challenge in well control and production optimization. The design of completions plays a significant role in improving the productivity of a well as well as the life of the reservoir. Intelligent or smart completions can be used to monitor, evaluate and manage the... 

The development of an oil field plays an important role in the optimum production. In the past, several wells have been used in a multi layer reservoir to produce oil from each layer. It costs oil industry billions of dollars and the problem would be more severe in the offshore. Nowadays, one well can produce oil from different pay zones in a reservoir. In this type of well called smart or intelligent well, production tubing is in common between pay zones and some chokes are installed on the tubing to produce oil from different layers. The size of choke usually varies and can be controlled by an electrical system installed in the well. As each pay interval has different capacity in terms of... 

Smart wellsareunconventional wells with down-hole devices installed on their production tubing. Such wells allow us to continuouslymonitor flow rates, pressures, temperature and periodically adjust down-holevalves setting without any production intervention. This makes smart well technology to be considered as one of the most significantbreakthroughs in reservoir management technologies. So, modeling and simulating smart wellsautomated action represents a challenge in forecastingtheir performances.One of the easiest ways to model such wells is to transform theminto a multi-segment well with the ability tocontrol each segment independently. By using the commercial reservoir simulator,... 

Upstream oil industry operators have turned to optimal production methods for further recoveries, such as smart wells. Control algorithms for smart wells must be equipped with online measurements, in other words, reflected in the feedback philosophy. On the other hand, after discretization, an oil reservoir's dynamic and control-driven model becomes a realization or a large-scale and sparse state space. One way to deal with this problem and provide an appropriate model for the design of smart well controllers is to use appropriate reservoir dimensional reduction methods. Therefore, to design low-order controllers, a study must first be performed at the level of the reservoir model and its...