Full text
| |
| Author(s): |
Carlos Eduardo Andrade Gomes Barreto
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | Campinas, SP. |
| Institution: | Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Mecânica |
| Defense date: | 2014-01-29 |
| Examining board members: |
Denis José Schiozer;
Rosângela Barros Zanoni Lopes Moreno;
Rubens Maciel Filho;
Ramiro Brito Willmersdorf;
Regis Kruel Romeu
|
| Advisor: | Denis José Schiozer |
| Abstract | |
The selection of number and position of inflow control valves in petroleum wells is complex and involves a high number of analyses to determine their viability of installation. Reservoir simulation can be used to represent valve operation and forecast production to evaluate the benefits in long terms. However, in specific cases, the high number of variables involved in this type of problem associated with the usual time-consuming of a single reservoir simulation made traditional optimization methods inefficient to solve the problem within an adequate time frame. This work proposes a dynamic optimization process that uses economic and technical indicators to speed up the process. The main ideia is to improve the exploration efficiency of the search space by selecting more potential well regions to start the process and to use an optimization workflow that splits the whole optimization in some suboptimization processes to evaluate each region at time. To assess the effectiveness of the proposed method, its results are checked by comparing them with those of an evolutionary algorithm using a simple example. The methodology is also applied in a more complex example with different geological scenarios. The results show that the proposed method achieves good results when compared with the evolutionary algorithm results. The design optimization in the complex example shows that the dynamic process is able to significantly increase the Net Present Value of the field with an acceptable number of simulation runs. It is also shown that the use of economic and technical indicators can be applied to reduce the number of variables of the problem, to define suitable constraints for each variable and to help the initial guess of seeds for the optimization method. It is concluded that the proposed methodology can be efficiently used to optimize inflow control valve design in cases in which computational resources and available time are limited. The use of indicators to better explore the search space is a viable way to selected more potential alternatives to be analyzed. The division of the whole optimization process in smaller pieces is an alternative to reduce the number of variables and to make faster evaluation, whitout loosing precision. In addition, the application of valves can improve total oil production, to retard water breakthrough and to reduce the water flow rate in the begginig of the water production, even with an increase in the total water production (AU) | |