生产数据的整合与初步分析在钻井中的应用实例
投稿时间:2021-05-28  修订日期:2021-05-28  点此下载全文
引用本文:
摘要点击次数: 67
全文下载次数: 0
作者单位邮编
李谦 成都理工大学环境与土木工程学院四川 成都 610059 610059
周长春 成都理工大学环境与土木工程学院四川 成都 610059 610059
朱海燕 成都理工大学能源学院四川 成都 610059 610059
鲁柳利 成都工业学院大数据与人工智能学院四川 成都 611730 611730
基金项目:中海石油(中国)有限公司项目“南海西部油田上产2000万方钻完井关键技术研究”子课题“乐东10区超高温高压井综合提速技术研究”(编号:CNOOC-KJ135ZDXM38ZJ05ZJ);四川省科技支撑计划应用基础研究项目“四川深层页岩气产能大数据挖掘和智能评估方法研究”(编号:2021YJ0360)
中文摘要:钻井过程中的生产数据是推动产业发展的重要驱动力,也是未来人工智能在钻井行业应用的基础。当前国内外行业巨头均已开始建立生产数据的收集与分析平台,但普通生产一线作业的数据整合与分析仍未引起重视。本文以采集自南海某区域10口井44种不同参数共21912条数据为例展现了生产数据从采集至定量分析的全流程。通过图像点采算法与数据插值补齐算法,实现不同格式数据的矩阵化整合。经过标准化与可视化的处理,可完成对整合数据的定性分析,明确生产数据的规律与趋势。在此基础上,通过统计分析、相关性分析与因子分析,可获得数据特征值,也能明确不同数据间的相互关系。分析结论实现了数据的分组与降维,在保证后续数据建模、人工智能等分析精度的同时降低了建模复杂度。
中文关键词:钻探数据  矩阵化整合  标准化  可视化  统计分析  相关性分析  因子分析  人工智能
 
Application of integration and preliminary analysis of production data in drilling
Abstract:The production data is an important driving force to promote the development of the drilling industry, and it is also the basis for the future application of artificial intelligence in the drilling industry. At present, all the drilling industry giants, either domestic or abroad, have begun to establish production data collection and analysis platforms. But the data integration and analysis for ordinary drilling production operations have still not attracted attention. The proposed paper takes 21912 data collected from 10 wells in a certain area of the South China Sea with 44 different parameters as an example to show the whole analyze process using production data, which from collection to quantitative analysis. The matrix integration of data in different formats can be realized from the image point sampling algorithm and data complementation algorithm. According to standardization and visualization processing, qualitative analysis of integrated data can be completed, and the law and trend of production data can be clarified. Based on the statistical analysis, correlation analysis and factor analysis, the data characteristic values can be obtained, meanwhile the interrelationship between different parameters can be clarified. Realizing the parameters grouping and dimensionality reduction, the accuracy of further data modelling can be ensured with the reduced modeling complexity.
keywords:drilling data  matrix integration  standardization  visualization  statistical analysis  correlation analysis  factor analysis  artificial intelligence
  查看/发表评论  下载PDF阅读器