الفهرس 
ContentsOnly 14 pages are availabe for public view
 |  | from | 109 |  |  |
| | | from | 109 | | |
Abstract
The increasing demand for oil all over the world has resulted in the exploitation of intelligent systems for the sake of reducing the uncertainty of exploring this important source. Finding the best well placement is associated with knowing the reservoir characteristics. These two problems are mainly the most important exploration problems to be optimized. The current study suggests a hybrid heuristic system that determines the optimum well location. The proposed system is developed to predicate porosity (reservoir characteristic) from seismic attributes then finds the optimal location of well depending on these predictions. The GANNMOG system is a hybrid system that uses three algorithms. Namely, Genetic Algorithm, Neural Network Algorithm, and Multi-Objective Genetic Algorithm. To evaluate the efficiency of the proposed GANNMOG system, we compared its performance with other models in current use. The GANNMOG system proved to be superior in solving well placement optimization. This thesis is organized in five chapters, as follows : Chapter one : Introduction and Objectives In this chapter, we review the goals of the thesis and the main purpose of the study. Chapter Two: Background and Challenges. This chapter has been prepared to provide an introduction to some of the basic concepts in the petroleum industry. The most important artificial intelligence algorithms that researchers used to solve exploration problems were also introduced. We explained how to benefit from the applications of artificial intelligence in the petroleum industry. Chapter Three: A Proposed Framewor. It offers the proposed mechanism for a system that can choose the best well placement. The system works in two stages :- The first stage: predicting the porosity value of the rock, that is, the ratio of voids between the rocks, which is an indication of the presence of oil reservoirs. The second stage: The best drilling location is chosen based on the predicted porosity value and the depth of the reservoir. Chapter Four: Experimental Results. This chapter discusses the results of applying the proposed system on the data of a petroleum site. The results showed the effectiveness of the system in solving the problem of predicting the porosity value and how it is possible to rely on those results in choosing the optimum well placement. Chapter Five : Conclusions and Future Work. In this chapter, conclusions are presented based on the proposed framework in chapter 3 and the results of implementation in chapter 4. To open new research directions in this scientific area future work is also provided.