Font Size: a A A

Research On Water Flooding Injection Allocation Methods Of Dense Well Pattern Based On Support Vector Machine(SVM)

Posted on:2017-04-15Degree:MasterType:Thesis
Country:ChinaCandidate:H LiFull Text:PDF
GTID:2271330488460323Subject:Oil and gas field development project
Abstract/Summary:PDF Full Text Request
As the first capacity building block of the second development project, after the reconstruction of well pattern and the restructuring of stratigraphic system, the development effect of the central west of Sa-Zhong development zone is improved. But the difference of development effect between different reservoirs is big. Through analysis, it can be realized that the well groups in this block still have the potential to improve recovery, and it also faces the issues such as the dispersion of remaining oil and the large number of reservoir layers, the high density of well pattern, and so on.In order to give full consideration to the effect of interlayer difference, reservoir relevance, reservoir pressure difference, material balance and energy balance on interval injection, express the nonlinear relationship between all the influencing factors and water injection more accurately, improve the prediction accuracy of interval injection, based on Support Vector Machine(SVM), layered water injection method system for water driving in dense well pattern is established. By studying the optimal-relevance well group with better water flooding effect, which obtained by combining reservoir engineering method, the grey correlation method and analytic hierarchy process(ahp), the calculation model of interval water injection can be established with SVM. The average error of this model is less than 3%, and the maximum error is less than 20%. By applying this model to the interval water injection prediction of 184 injection wells(715 water injection interval)in Sa-Zhong area, the water-flooding adjustment scheme of this area was received. By analyzing the results of numerical simulation, we obtained that, after adjustment, the water cut of the area dropped for 0.247%, water cut rising velocity was reduced, and the oil production increased by 3.8702×104 tons. Therefore this model has enough precision and the utilization value.The injection allocation system of the injection well and its best effective production well group in this article, can be extended to the whole production of injection pattern of the central west capacity block, and the injection allocation method that are suitable for well group under the condition of dense well pattern can be achieved, which can guide the next step of water flooding development adjustment. It also has important guiding significance for other injection allocation research and application under the condition of dense well pattern.
Keywords/Search Tags:interval water injection, injection allocation methods, SVM(support vector machine), dense well pattern, correlation between oil and water wells
PDF Full Text Request
Related items