Automatic Identification of Gas Hydrate Formation using Machine Learning Algorithms（利用机器学习算法自动识别天然气水合物的形成）
发布时间：2024-01-08
元数据
名 称
Automatic Identification of Gas Hydrate Formation using Machine Learning Algorithms（利用机器学习算法自动识别天然气水合物的形成）
科技资源标识
CSTR:11738.14.NCDC.XDA14.PP6139.2024
数据共享方式
开放下载
摘 要
Ensemble learning integrates the results of multiple weak learners through certain rules, so it has better learning effect than single weak learner. As a method of machine learning, it has been widely used in various fields. Existing hydrate recognition methods are highly empirical and lack of efficiency. Gas hydrate has the characteristics of high resistivity, abnormal potential and low gamma parameters, so it can be considered to be identified by using machine learning method with well logging parameters. Compared with traditional hydrate recognition methods, ensemble learning has the characteristics of fast speed and high accuracy, and has a wide application prospect. In this paper, a complete data analysis process and corresponding data analysis algorithm are used to analyze and process logging data, and the accuracy of integrated learning in hydrate recognition is verified by comparing with the real formation conditions. Comparing different ensemble learning algorithms for the accuracy of final prediction results, the ensemble learning algorithm presents the highest accuracy, which can be applied for further analysis, e.g. dimension reduction and data training, thus reducing the workload of data processing.
宁禹强,叶智慧,王涵,陈冬,李守定,朱丹丹. Automatic Identification of Gas Hydrate Formation using Machine Learning Algorithms（利用机器学习算法自动识别天然气水合物的形成）. 国家冰川冻土沙漠科学数据中心(http://www.ncdc.ac.cn), 2024. https://cstr.cn/CSTR:11738.14.NCDC.XDA14.PP6139.2024.
本作品采用
知识共享署名
4.0 国际许可协议进行许可。
CSTR:11738.14.NCDC.XDA14.PP6139.2024