Forecast model for gas well productivity based on PSO and SVM

详细信息    查看全文 | 下载全文 | 推荐本文 |

• 作者：Min YangJun LiJingcheng Liu
• 会议时间：2011-11-01
• 关键词：gas well productivity ; PSO ; SVM ; forecast model
• 作者单位：Min Yang(Chongqing University of Science and Technology, Chongqing, China,401331)Jun Li(Sinopec Northwest Oilfield Branch Yakela gas plant,Xinjiang, China,842017)Jingcheng Liu(Chongqing University of Science and Technology, Chongqing, China,401331 ;Chongqing Petroleum And Natural gas Society,Chongqing,China,400030)
• 母体文献：2011全国特殊气藏开发技术研讨会论文集
• 会议名称：2011全国特殊气藏开发技术研讨会
• 会议地点：重庆
• 主办单位：重庆市科学技术协会
• 语种：chi

摘要

It is very important to forecast the gas well productivity of gas reservoir accurately.On the basis of analyzing the parameter performance of support vector machine (SVM) for regression estimation, the paper proposes gas well productivity prediction model based on particle swarm optimization (PSO) and SVM.The parameter of SVM was optimized by PSO.This method took advantage of the minimum structure risk of SVM and the quickly globally optimizing ability of PSO.Compared with BP neural network model, the proposed GA-SVM model for gas well productivity in practical engineering has higher accuracy and speed, and the maximum error is 2.8%.Thus, it provided a new approach to predict the gas well productivity.