#TITLE_ALTERNATIVE#

• Main Author: (NIM : 12212014), IRWAN
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/28145
• Institution: Institut Teknologi Bandung

Optimizing production is no longer an option, nowadays, it is a necessity. The wells have to optimized more economically than before in order to influence projects attractiveness. The positioning of well is one of the most important aspect in production strategy plan and optimization and is highly complex. The problem become more complex when horizontal wells and its interactions are considered. Solving this problem is time-consuming and demands high computational effort. Therefore, the objective of this research is to provide an optimization algorithm resulting good solution efficiently. <br /> <br /> The optimization algorithm used in this work is the hybrid genetic algorithm (HGA) which is the combination of GA with artificial neural networks (ANN) and evolution strategies (ES). This HGA attempts to simplify the complex and diverse parameters governing the horizontal well optimization problem. The HGA is coupled with commercial simulator and tested in real field model to quantify the benefits of this HGA over a base case with the conventional one. The algorithm in this study is handcrafted specially for oilfield industry so that the number of individuals in a population, the number of generations, and the genetic parameters are varied and adapted by applying ANN and ES to eliminate the need to tuning strategy parameters thus bridge the gap between petroleum engineer and algorithm engineer. <br /> <br /> The HGA proposed in this study is used to optimize the horizontal well parameters in realistic reservoir model defining wellhead location, target layer, horizontal section length, azimuth angle, and producing rate. The HGA is proved to be suitable for any kind of reservoir models if the benchmark model for optimization approves. <br /> <br /> There are 3 cases of horizontal well development plan which will be discussed, in the first case two wells are added at the same time, in the second case, two wells are added at different time with 3 years gap, and in the third case, is basicly second case with optimum rate considered. <br /> <br /> Result comparison between performance of HGA and conventional method is discussed. The result pronounce HGA are superior than conventional method in optimizing horizontal well in all three cases tested. The superiority is increasing with a more complex development plan problem between these three. The supplementary explanation is included to give understanding why HGA is superior over the conventional method in optimizing horizontal well. <br /> <br /> Based on the fact that every horizontal well have different challenges that defines a single-solution approach which results the horizontal well optimization becoming more and more complicated, the use of an optimization algorithm to achieve a best solution is very valuable to the process, yet it can also lead to an exhaustive search, demanding a great computation power to test many possibilities. This novel HGA is presented to tackle and handle those issues.