School of Energy Resources,China University of Geosciences (Beijing) Coal Reservoir Lab,National Engineering Research Center of Coalbed Methane Development and Utilization

为了准确选择不同生产类型煤层气井的产能预测模型,根据数理统计理论,并通过实例分析讨论了基于时间序列的翁氏模型、月产/累产比模型及灰色系统模型在煤层气水平井的产能拟合和预测中的有效性、适用性及差异性。研究表明:3种模型均能对煤层气井的生产历史进行很好地拟合,并进行准确预测,但又各有差别。翁氏模型对产量变化平稳、波动小的井预测准确,波动大时预测误差大;月产/累产比模型可进行中长期预测,但受递减规律的影响;灰色系统模型对生产井递减阶段拟合度及预测精度高,但误差随递减数据的波动变大。因此,应针对煤层气井的生产阶段、生产特征和目的,有选择地应用。

in order to accurately choose suitable productvity prediction model for different production type coalbed methane wel.according to the mathematical statistic theory and case analysis,the paper discussed the effiency,sutability and dfference ofthe Weng model,.month producion/accumulated production ratio model and gray system model based on the time sequence in the iting and prediction of coalbed methane horizontal wels.The study showed that the above three models could all well fiting on the production history of the coalbed methane wels could make the accurate predictions and could have differences in each model.The Weng model could make an accurate prediction for the wells with a stable variaton production and a smalfuctuation and when a high fuctuation occurred,the prediction would have a high eror.The month production/accumulated production ratio model couldbe applied to the middle-long term prediction and would be affected by the decreasing lavw.The gray system model would have a igh fiting and high prediction accuracy to the decreasing stage of the production wells and the eror would be increased with the fluctuation of the decreased data.Therefore,he models should be selectively applied according to the production stage,production features and purpose of the coalbed methane well.

国家自然科学基金资助项目(41272175)；国家自然科学基金青年基金资助项目(41502157)；