中国石油大学学报（自然科学版）2026，Vol.50Issue(2)：64-73,10.DOI:10.3969/j.issn.1673-5005.2026.02.007
基于Granger-LSTM模型的东营凹陷页岩油产量预测
Shale oil production prediction in Dongying Depression based on Granger-LSTM model
摘要
Abstract
The production dynamics of shale oil horizontal wells are highly complex,and existing prediction techniques often fail to achieve satisfactory accuracy.In this study,production data from wells X1 and X2 in Dongying Depression were ana-lyzed.Granger causality analysis was first employed to identify time-varying factors strongly correlated with shale oil produc-tion,thereby optimizing the model input features.Subsequently,a long short-term memory(LSTM)network was developed to construct the production prediction model,with its hyperparameters optimized using the particle swarm optimization(PSO)algorithm.The predictive performance of the LSTM model was compared with recurrent nural ntworks(RNN),gted rcurrent uits(GRU),and temporal cnvolutional ntworks(TCN).The results highlight the critical role of feature selection in produc-tion forecasting.For example,for well X1,the LSTM model incorporating Granger-based features reduced the root mean square error by 3.41 m3 compared with the model using features selected via Spearman analysis,significantly enhancing pre-diction accuracy.Although various time-series models exhibit strong predictive capabilities,the LSTM model outperforms others in capturing dynamic temporal characteristics,providing a solid theoretical basis and technical support for complex shale oil production forecasting.关键词
页岩油产量/东营凹陷/Granger分析/长短期记忆网络(LSTM)/粒子群优化/预测模型Key words
shale oil production/Dongying Depression/Granger analysis/long short-term memory(LSTM)/particle swarm optimization(PSO)/prediction model分类
天文与地球科学引用本文复制引用
张凤姣,张晋言,邓少贵,齐国华,范中专,孙鑫..基于Granger-LSTM模型的东营凹陷页岩油产量预测[J].中国石油大学学报（自然科学版）,2026,50(2):64-73,10.基金项目
国家自然科学基金项目(42074134) （42074134）
