平均模型在致密气井产量预测中的应用

在致密气资产快速评价中应用经验型递减模型能够快速得到气井产量的递减情况;但不同的递减模型在完成历史拟合后,长期产量和累积产量预测结果相差很多。应用GLUE法评价各模型预测结果的不确定性,并引入平均模型的概念,计算各模型权重后建立平均模型。以加拿大致密气田两口致密气井的实际生产数据为例应用平均模型进行累积产量的计算,结果表明GLUE法可进行各模型预测结果的不确定性进行分析,且通过设定筛选标准后,利用满足条件的模型进行聚合得到的平均模型计算结果具有更小的标准差与变异系数。

Application on Tight Gas Production with Average Method

During the tight gas assets rapid assessment, empirical decline model can quickly get the production decline prediction of tight gas well. Each model can get perfect history matching, but the production forecast of each model is very different. Use GLUE method to analysis the uncertainty of each prediction model and calculate the weights of each model to create an average method. The production data of two tight gas well in Canada was analyzed to show the method, and results show that the GLUE method can be used to analyze uncertainty of the model prediction results and through the screening standard setting, with satisfy the standard model are combined to get the average model, the average model calculation result has a smaller standard deviation and coefficient of variation.