Applying an optimized proxy-based workflow for fast history matching
- PDF / 2,237,312 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 0 Downloads / 192 Views
ORIGINAL PAPER
Applying an optimized proxy-based workflow for fast history matching Mojtaba Karimi 1 & Ali Mortazavi 1 & Mohammad Ahmadi 1
Received: 6 June 2017 / Accepted: 16 October 2017 # Saudi Society for Geosciences 2017
Abstract History matching is still one of the main challenging parts of reservoir study especially in giant brown oil fields with lots of wells. In these cases, history matching with conventional manual technique needs many runs and takes months to get a match. In this work, an innovative approach was suggested for fast history matching in a real brown field. The workflow was employed based on an optimized proxy model for history matching of a field consisting of 14 active wells with multiple responses (which are production rate and pressure data) in the south part of Iran. The main important features of the proposed algorithm were defining a proxy model which is response surface method in which 21 model parameters were incorporated based on cubic centered face method. The proxy model was then optimized by one of the most famous algorithms which is genetic algorithm. Proxy model was successfully performed using 256 samples leading into p- value of 0.531 and R2 of 0.91 dataset. As a result, the proposed workflow and algorithm showed good and acceptable results for history matching of studied real model.
Keywords Response surface method . Genetic algorithm . Fast history matching . Proxy model
Introduction Reservoir full field study usually takes a long time, and shortening this time without sacrificing work quality is one of the objectives of reservoir management especially in mature and giant oil and gas fields. There is no unified single method for history matching in reservoir analysis since the reservoir problems are highly site and case specific. Geological conditions, history, drive mechanisms, and well specifications are properties that are unique for each field (Shahkarami 2012). The causes of long time associated with conventional history matching in reservoir study can be categorized into three main groups: a) Data gathering, property modeling, and reservoir simulation b) Prehistory matching which includes model checking, properties, and responses quality check c) History matching In this paper, after checking the important factors of real mature fields, such as maps, permeability, relative permeability, petrophysics, aquifer properties, PVT, and static/ geological and dynamic models, a computer-based workflow was introduced to accelerate the history matching speed in a mature field consisting of some wells.
* Ali Mortazavi [email protected]
Background Mojtaba Karimi [email protected] Mohammad Ahmadi [email protected] 1
Amirkabir University of Technology, Tehran, Iran
Several studies have defined a methodical style to execute history matching (Al-Thuwaini et al. 2006; Amini et al. 2014; Arief 2013; Bhark and K 2014; Cancelliere et al. 2013). Sultan et al. (1994) used the simulated annealing to perform history matching process (Ouenes et al. 1993;
462
Arab J Geosci
Data Loading...
