基于元建模的裂缝性气藏累计产量预测

为了预测某裂缝性气藏在18年规划周期内的累计产量,建立了3种元模型,并结合数值模拟累计产量预测结果对各模型进行了验证和对比。估算了气藏的基本参数(孔隙度、渗透率及含水饱和度)并对气藏进行了数值模拟。将气藏内某井18年内的累计产量表示为各气藏参数的函数,采用储量计算经验函数排除了研究空间内的低潜力点,并基于最大熵准则选取25个高潜力点作为设计点。建立了二次模型、乘性模型和RBF模型3种元模型,基于各模型预测了25个设计点和7个试验点处的累计产量。预测结果表明:各模型均能比较准确地预测累计产量;RBF模型的累计产量预测准确性优于另外2种模型;与数值模拟方法相比,元建模方法的计算时间大幅缩短。

Prediction of accumulated outflow from a fractured hydrocarbon reservoir using metamodeling

Three metamodels were established for predicting accumulated outflow from a fractured hydrocarbon reservoir over a planning horizon of 18 years, and the models were validated and compared using accumulated outflow predicted by numerical simulation. The reservoir was simulated and its basic parameters (porosity, permeability and water saturation) were estimated. The accumulated outflow over 18 years of a well in the reservoir was expressed as a function of the reservoir parameters. Low potential points were excluded using HIP (Hydrocarbon in Place) equation and 25 high potential points were chosen as design points using maximum entropy design. Three kinds of metamodels (quadratic model, multiplicative model and radial basis function model) were built and the accumulated outflows of 25 design points and 7 test points were predicted based on the models. The prediction results show that all of the three models can accurately predict the accumulated outflow of the reservoir studied in this paper and the radial basis function model outperforms the other two metamodels. Besides, the calculating time of the metamodeling method is much less than that of the numerical simulation.