摘要
为了探索云计算与人工智能技术在地震资料解释领域的结合方式,更好地发挥两者的协同优势,在深度学习解释软件研发实践的基础上,提出了基于梦想云+深度学习方法的地震资料解释新模式:①由梦想云数据湖与计算中心为深度学习算法提供大量的高质量标签数据与高弹性计算平台;②深度学习技术与传统地球物理方法优势互补,共同搭建地震与地质解释的新流程;③在梦想云平台上应用深度学习模块,直接为梦想云平台的业务决策场景提供有效地震成果信息支持。这种模式可以取代大量手工操作、实现更高预测精度,从而为高效益、高精度油气勘探打下坚实的基础,经实践验证,取得了令人鼓舞的效果。
In order to explore the combination of cloud computing and artificial intelligence technology in seismic data interpretation, and to give play to their collaborative advantages, on the basis of the R&D practice of deep-learning seismic interpretation software, a new seismic data interpretation mode based on E&P Dream Cloud and deep learning method has been proposed, which includes:(1) Data Lake and Computing Center of E&P Dream Cloud provide a large number of high-quality label data and high elastic computing platform for the deep learning algorithm.(2) Deep-learning technology and traditional geophysical methods complement each other to establish a new process for seismic and geological interpretation.(3) The application of deep-learning module on the E&P Dream Cloud platform directly provides effective information support of seismic achievements for business decision-making on E&P Dream Cloud platform. This mode can save a large number of manual operations and achieve higher prediction accuracy, thus laying a solid foundation for high-efficiency and high-precision oil and gas exploration. After real data practice, encouraging results have been achieved.
作者
杨平
詹仕凡
李明
李磊
郭锐
尚民强
陶春峰
Yang Ping;Zhan Shifan;Li Ming;Li Lei;Guo Rui;Shang Minqiang;Tao Chunfeng(BGP Inc.,China National Petroleum Corporation)
出处
《中国石油勘探》
CAS
CSCD
北大核心
2020年第5期89-96,共8页
China Petroleum Exploration
基金
中国石油天然气集团有限公司科学研究与技术开发项目“深层与非常规物探新方法新技术”(2019A-33)
中国石油天然气股份有限公司投资信息化重点项目“勘探开发一体化协同研究及应用平台(一期)建设”(PetroChina-IT-2017-N104)
中国石油集团东方地球物理勘探有限责任公司科技项目“储层高分辨地震反演技术深化研究及软件集成”(01-03-02-2019)。
关键词
梦想云
深度学习
人工智能
地震资料解释
E&P Dream Cloud
deep learning
artificial intelligence
seismic data interpretation
