底青云,李守定,付长民,等. 2021.基于云端大数据的智能导向钻井技术方法[J].工程地质学报,29(1):1178-1185. doi:10.13544/j.cnki.jeg.2021-0055. DOI: 10.13544/j.cnki.jeg.2021-0055
    引用本文: 底青云,李守定,付长民,等. 2021.基于云端大数据的智能导向钻井技术方法[J].工程地质学报,29(1):1178-1185. doi:10.13544/j.cnki.jeg.2021-0055. DOI: 10.13544/j.cnki.jeg.2021-0055
    Di Qingyun, Li Shouding, Fu Changmin, et al. 2021. Intelligent steering drilling technology method based on cloud big data[J]. Journal of Engineering Geology, 29(1): 1178-1185. doi: 10.13544/j.cnki.jeg.2021-0055.
    Citation: Di Qingyun, Li Shouding, Fu Changmin, et al. 2021. Intelligent steering drilling technology method based on cloud big data[J]. Journal of Engineering Geology, 29(1): 1178-1185. doi: 10.13544/j.cnki.jeg.2021-0055.

    基于云端大数据的智能导向钻井技术方法

    INTELLIGENT STEERING DRILLING TECHNOLOGY METHOD BASED ON CLOUD BIG DATA

    • 摘要: 导向钻井技术方法是21世纪全球石油工业最重要的技术之一,也是美国“页岩气革命”核心技术水平钻井的关键组成部分。当前,导向钻井的主要研究目标是提高钻井速度、降低钻井时间和风险,智能化是目标实现的重要途径。文章分析了国内外大数据与人工智能在石油工业应用情况,建立了云端大数据智能导向钻井方法架构,提出了随钻测井参数人工智能反演与识别方法,指出了云端大数据与智能算法管理的实现途径,得出如下结论:(1)基于云端大数据智能导向钻井方法主要包括物联网感知层、大数据存储层和云平台决策层。物联网感知层实现井场关键信息的采集并传输至大数据中心;大数据中心支持数据存储与云管理;云平台决策层依托大数据中心的海量数据,进行云端地面软件控制、人工智能决策以及云平台管理。(2)采用机器学习的方法智能反演与识别地层岩性,选择自然电位、自然伽马、密度、声波、补偿中子、电阻率等6条随钻测井数据,分别采用不同的机器学习算法进行地层岩性反演与识别,决策树模型和随机森林模型分别达到0.81和0.89的准确度,形成了一套可快速自动描述岩性特性分类的方案。(3)云端平台管理决策可进行井下实时数据解码,获取钻井轨迹和测井曲线,其中云端人工智能决策模块对地层及钻井参数进行智能反演预测,可实现钻井轨迹智能修正和钻井参数智能优化,保证智能导向工程钻得准、钻得快。

       

      Abstract: Steering drilling technology is one of the most important technologies in the global petroleum industry in the 21st century, and it is also a key component of horizontal drilling, the core technology of the American "The Shale Gas Revolution". At present, the main research goal of steering drilling is to increase the drilling speed, reduce the drilling time and risk, and intelligence is an important way to achieve this goal. The article analyzes the application of big data and artificial intelligence in the petroleum industry at home and abroad, establishes a cloud big data intelligent steering drilling method framework, proposes an artificial intelligence inversion method for logging while drilling parameters, and points out the way to realize the management of cloud big data and intelligent algorithms, and draw the following conclusions: (1)The intelligent guided drilling method based on cloud big data mainly includes the things perception layer, the big data storage layer and the cloud platform decision layer. The things perception layer realizes the collection and transmission of key information of the wellsite to the big data center. The big data storage center is mainly responsible for data storage and cloud management. The cloud platform decision layer relies on the massive data in the big data center to perform cloud ground software control, artificial intelligence decision-making, and cloud platform management. (2)Select six geophysical parameters such as SP, GR, DEN, AC, CNL, and RT, and use different Machine Learning algorithms to build models to realize the independent identification of formation lithology. The Decision Tree model and the Random Forest model have an accuracy of 0.81 and 0.89 respectively, forming a set of schemes that can quickly and automatically describe the classification of lithological characteristics. (3)The cloud platform management decision is mainly used to decode real-time upload data downhole, and obtain drilling trajectories and logging curves. The cloud artificial intelligence decision-making module performs intelligent inversion and prediction of stratum and drilling parameters, realizes intelligent correction of drilling trajectories and intelligent optimization of drilling parameters, and ensures the accuracy and speed of drilling of intelligent steering engineering.

       

    /

    返回文章
    返回