Font Size: a A A

Three Kinds Of Abroad Commonly Used Analytical Model Of Steam Driving Analysis And Its Applicability Study

Posted on:2016-03-29Degree:MasterType:Thesis
Country:ChinaCandidate:Y FuFull Text:PDF
GTID:2271330461481365Subject:Oil and gas field development project
Abstract/Summary:PDF Full Text Request
Heavy oil occupy large proportion in the oil and gas resources in the world. Accounted for about 28% of the total petroleum resources in our country. Heavy oil exploitation plays an important role in sustainable development in the entire oil industry in our country. Only for steam stimulation of heavy oil reservoir exploitation, oil recovery was only 10-20%, remaining bypassed oil is too much, the remaining reserve is too big, should be carried out and then steam flooding, to increase the recovery efficiency to more than 50%.At present, there are some numerical simulation software can be used to steam drive development and forecast, although the prediction effect is good, but the software itself is not cheap, often need to apply a large computer, in addition will also spend a lot of computing time. Therefore substitute analysis prediction model for numerical simulation software also can be regarded as a good method. Steam drive displacement process is difficult to analyze. The same prediction model for different reservoir types, different geological conditions of oil field prediction effect is different. The mechanism of steam driving analysis and study of prediction model is necessary, for heavy oil production practice have important guiding significance. This paper comprehensive comparison of different steam driving model, selected the abroad nowadays the most commonly used three kinds of steam drive model, in-depth analysis and study of the three models to the applicability of different reservoir types.Separately thorough research the model of Jones,Farouq Ali,Miller-Leung. The computer program was compiled. Through history matching, found that Jones steam driving model prediction effect is better on Kern River oil field and similar oilfield, stronger applicability. On other types of oilfield productivity prediction need to input some experience parameters and do some adjustments to the history matching. In prediction of oil production rate, Jones model at the beginning of the production is low, in the later time and prediction on the high side. On the forecast of cumulative production, at the start of prediction accuracy is not high, but as time goes on, the prediction results will be more and more accurate. The revised Farouq Ali model does not need any experience parameters can be fitted well. But model requires Sors, Sorw, Sst and Swir parameters is difficult to get, need a brief estimate computing. Comparing Farouq Ali model with the ideal model in CMG numerical simulation software, can know the adjusted Farouq Ali model is more fitting Kern River oilfield. Miller-Leung model introduces the concept of area spread coefficient, increase the scope of the model. In the prediction of oil production rate, Miller-Leung model curve more smoothly, by contrast, found the forecast of oil cumulative production, especially the application of large oil block, Miller-Leung model than the Jones model is more accurate and good, but for individual oilfields, due to the influence of some factors, Miller-Leung model predicting time is too short.Through the steam injection temperature, net total thickness ratio, initial oil saturation and a series of sensitive parameters study, know which parameters has great influence, little influence, no influence on oil field production, even can target specific reservoir conditions adjust the steam injection parameters in the production.
Keywords/Search Tags:steam driving, cumulative production prediction, History matching, Sensitive parameters, computer program
PDF Full Text Request
Related items