Combination prediction and error analysis of conventional gas production in Sichuan Basin

Author:

Li Haitao,Yu Guo,Chen Yanru,Fang Yizhu,Chen Yu,Zhang Dongming

Abstract

The accurate prediction of the trend of natural gas production changes plays an important role in the formulation of development planning plans. The conventional gas exploration and development in Sichuan Basin has a long history. Based on the development of conventional natural gas production, the article uses the Hubbert model, Gauss model, and GM (1, N) model to predict conventional natural gas production, and then the Shapley value method is used to allocate the weight values of the three models, and a combination model for conventional gas production prediction is established. Finally, residual analysis and precision test are carried out on the prediction results. The results show that: 1) The combination model established using the Shapley value method can effectively combine the advantages of various models and improve the accuracy of prediction. And the standardized residual of the combined model is the lowest, the prediction is closest to the actual value, and the accuracy test is the best, indicating that the combined model has the highest accuracy. 2) After using a combination model for prediction, conventional gas production will peak in 2046, with a peak production of 412 × 108 m3, with a stable production period of (2038–2054) years, a stable production period of 17 years, and a stable production period of 389 × 108 m3, the predicted results of the combined model have a longer stable production period, and the trend of production changes is more stable. The use of combination model provides a reference for the field of natural gas prediction, while improving the accuracy of prediction results and providing better guidance for production planning.

Publisher

Frontiers Media SA

Subject

General Earth and Planetary Sciences

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3