[1]李春生,陈思宇,张可佳,等.勘探开发数据智能质量检查方法研究[J].计算机技术与发展,2022,32(08):180-184.[doi:10. 3969 / j. issn. 1673-629X. 2022. 08. 029]
　LI Chun-sheng,CHEN Si-yu,ZHANG Ke-jia,et al.Research on Intelligent Quality Check Method of Exploration and Development Data[J].,2022,32(08):180-184.[doi:10. 3969 / j. issn. 1673-629X. 2022. 08. 029]
点击复制

勘探开发数据智能质量检查方法研究()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年08期
页码:
180-184
栏目:
应用前沿与综合
出版日期:
2022-08-10

文章信息/Info

Title:
Research on Intelligent Quality Check Method of Exploration and Development Data
文章编号:
1673-629X(2022)08-0180-05
作者:
李春生; 陈思宇; 张可佳; 富 宇; 刘 涛
东北石油大学 计算机与信息技术学院,黑龙江 大庆 163318
Author(s):
LI Chun-sheng; CHEN Si-yu; ZHANG Ke-jia; FU Yu; LIU Tao
School of Computer and Information Technology,Northeast Petroleum University,Daqing 163318,China
关键词:
勘探开发数据; 智能质检; 专家系统; 机器学习; 异常检测
Keywords:
exploration and development data; intelligent quality inspection; expert system; machine learning; anomaly detection
分类号:
TP182
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 08. 029
摘要:
油田勘探开发数据库作为油田的重要数据分析资源,其数据的准确性受到油田工作人员的广泛关注。 随着勘探开发数据的不断增加,数据质检难度逐渐增大,但现存的质量检查方法不能及时、有效地发现数据质量问题。 现存的质检方法是管理人员凭借个人经验并在数据库中建立存储过程进行数据检查,此方法存在以下两个方面的问题:一方面工作效率低且工作人员在统计数据时容易出现错误输入、数据遗漏等情况,导致结果准确性低,另一方面存储过程包含的质检规则宽松、灵活性差、缺少智能性,仅仅对数据的完整性、一致性进行检查,只能检查出部分异常数据。 现急需一种智能化的质检方法来解决当前质检速度慢、难度大的问题。 因此,提出建立质检专家系统,并使用机器学习算法辅助使其智能地完成质检工作,减轻数据管理人员的压力。 该文通过采集专家经验建立知识库,而后使用推理机推理得到质检结果,再应用机器学习算法分析质检结果对知识库进行更新,最后通过实验证明该方法可以有效地对勘探开发数据进行质量检查。
Abstract:
Oilfield exploration and development database is an important data analysis resource,and its data accuracy is widely concernedby oilfield staff. Along with the increase of exploration and development data,the data quality control difficulty increases gradually,butthe existing quality inspection method can’t timely and effectively discover data quality problems. The existing? quality inspection methodis that the management personnel rely on personal experience and establish stored procedures in the database for data inspection. Thismethod has the following two problems. On the one hand,the work efficiency is low and the staff are prone to error input and data omission in the statistical data,resulting in low accuracy of the results. On the other hand,stored procedures contain loose quality inspectionrules with poor flexibility and lack of intelligence. They only check the integrity and consistency of data,and only some abnormal datacan be detected. An intelligent quality control method is urgently needed to solve the problems of slow speed and difficulty in qualitycontrol. Therefore,a quality inspection expert system? is proposed,and the machine learning algorithm is used to assist it to complete thequality inspection work intelligently,so as to reduce the pressure of data management personnel. In this paper,the knowledge base is established by collecting expert experience,and then the inference machine is used to deduce the quality inspection results. Then the machine learning algorithm is applied to analyze the quality inspection results and update the knowledge base. Finally,experiment shows thatthe proposed method can effectively check the quality of exploration and development data.
更新日期/Last Update: 2022-08-10