福建电脑2025，Vol.41Issue(3)：16-19,4.DOI:10.16707/j.cnki.fjpc.2025.03.004
TCN-DBN模型在油井产油量预测中的应用
Application of TCN-DBN Model in Oil Production Prediction of Oil Wells
摘要
Abstract
This paper proposes a TCN-DBN algorithm for oil well production prediction to address the issues of parallelism and gradient stability in long short-term memory neural networks.Combining temporal convolutional networks with deep belief networks,the algorithm uses temporal convolutional networks to extract temporal series features,while integrating self attention mechanisms to improve model prediction accuracy.The experimental results showed that compared with the LSTM-DBN algorithm,the TCN-DBN algorithm reduced the MAE value by 72.95%,RMSE value by 68.94%,and MAPE value by 48.47%on the Z25 production well.The experimental results demonstrate that the TCN-DBN algorithm has high prediction accuracy and can provide theoretical and technical support for the formulation of crude oil extraction plans.关键词
时序卷积网络/深度置信网络/油井产油量预测Key words
Temporal Convolutional Network/Deep Belief Network/Oil Well Production Prediction分类
石油、天然气工程引用本文复制引用
潘少伟,宋倩,王树楷..TCN-DBN模型在油井产油量预测中的应用[J].福建电脑,2025,41(3):16-19,4.基金项目
本文为西安石油大学2023年度"立德树人"研究课题"OBE理念下思政元素融入程序设计类课程教学的探索与实践—以《JAVA程序设计》为例"(No.LD202309)的研究成果,并获得它的资助. （No.LD202309）
