Unsupervised Learning for Oil Drilling Data Classification

Isam, Sherrif, Yan, Lipeng and Zhang, Qichun (2021) Unsupervised Learning for Oil Drilling Data Classification. In: 2021 26th International Conference on Automation and Computing, 02-09-2021 Through 04-09-2021.

Full text not available from this repository. (Request a copy)

Abstract

In this paper, data-driven classification methods has been adopted for oil drilling process to describe the formation information. In particular, the unsupervised machine learning algorithms, including PCA, k-mean, t-SNE, have been used to deal with the experimental well log data. Based on the results, the clusters will reflect the range of the geologic interpretation. Basically, 9 clusters have been obtained following PCA and k-mean while 6 clusters have been formed using t-SNE. Note that the merged 3 clusters are very small in scale then the results are consistent in terms of value ranges, while the optimal number of the cluster has been further discussed as 3 clusters by elbow method. It has been shown that the geologic information can be obtained via data-based analysis directly.

Item Type:	Conference or Workshop Item (Paper)
Depositing User:	RED Unit Admin
Date Deposited:	27 May 2025 13:40
Last Modified:	27 May 2025 13:40
URI:	https://bnu.repository.guildhe.ac.uk/id/eprint/20263

Actions (login required)

Edit Item