Application of Evolutionary Algorithm for Assisted History Matching
History matching is a fundamental technique in reservoir engineering principle. Successful reservoir interpretation mostly depends on the precision of the history matching. History matching is an act of adjusting the developed model in simulating the past reservoir performance to match the actual hi...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Year Project |
| Language: | English |
| Published: |
Universiti Teknologi PETRONAS
2014
|
| Subjects: | |
| Online Access: | http://utpedia.utp.edu.my/14224/1/DISSERTATION%20REPORT%20%282%29.pdf http://utpedia.utp.edu.my/14224/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | History matching is a fundamental technique in reservoir engineering principle. Successful reservoir interpretation mostly depends on the precision of the history matching. History matching is an act of adjusting the developed model in simulating the past reservoir performance to match the actual historical data. From the outcome, engineers are able to estimate the future production rate of the well closely based on parameters like pressure, relative permeability and porosity. When the differences between the observed performance data and simulated data are found, the iterations are made to modify the accuracy of the match. Traditionally, this iterative technique is computed manually which is very time consuming.
The development of history matching technique has evolved rapidly over the past 20 years from manual to automated history matching. As the technology moving on, history matching is also improvised in scope of optimization. Generally, history matching consists of manual and automatic computation. Manual execution commonly apply trial-and-error concept which the probability ranges is quite uncertain and time consuming. Besides, it really demands skill and experience on the part of simulation engineer. Today, tremendous efforts are made to develop Automatic History Matching algorithms. While the automatic method focus on optimization which is normally computer based.
In this project, we will define and discuss the application of evolutionary algorithm in assisted history matching. Evolutionary method helps to find the global minima directly without the presence of local minima. Besides, algorithm based method has been widely used to forecast future result in various field for example art, biology, marketing including engineering. The methodology will be tested on developed synthetic model. |
|---|