预测页岩气产量递减组合模型的研究

Combination model for shale gas production decline prediction

刘传斌¹ 姜汉桥¹ 李俊键¹ 糜利栋¹ 赵林¹ 乐雪霖²

中国石油大学（北京）石油工程教育部重点实验室，北京 102249 中国石油川庆钻探工程有限公司四川石油天然气工程建设有限责任公司重庆分公司，重庆 400021

MOE Key Laboratory of Petroleum Engineering, China University of Petroleum, Beijing 102249, China Chongqing Company, Sichuan Petroleum Engineering Construction Corp. Ltd., Chuanqing Drilling Engineering Company Ltd., CNPC, Chongqing 400021, China

页岩气由于微￣纳米尺度流动机理和多尺度渗流规律的复杂性，给其产能评价和递减规律的预测带来很大挑战。针对Arps模型在页岩气产量递减预测中存在的局限性，学者们先后提出了SEPD模型、Duong模型和YM￣SEPD模型，对页岩气产量预测起到了一定的指导作用。文中通过数值模拟对比了这3种模型的适用性，结果表明，在生产后期SEPD模型和YM￣SEPD模型预测一般较生产数据偏低，Duong模型预测一般偏高。综合考虑这些模型的优缺点，提出了优化组合模型。根据组合模型的思路，生产20 a末的预测误差可控制在10%左右，较之前单一模型预测50%左右的误差，准确度大幅提高，并由实例应用进行了验证。

The complexity of micro and nano-scale flow mechanism and multi-scale seepage law presents great challenge to the shale gas production evaluation and decline prediction. In order to solve the limitation of Arps model, experts have proposed three major models(SEPD model, Duong model and YM-SEPD model), which have significance on shale gas production decline. Compared with the suitability of three models by numerical simulation, SEPD and YM-SEPD predictions are generally lower than the production data, while Duong model prediction generally is high. Summarizing the advantages and disadvantages of the three models, optimization combination models are proposed. Numerical simulation confirms error of combination model prediction can be controlled at around 10%, compared with the 50% error of single model, which has a substantial increase in accuracy.

10.6056/dkyqt201504016