Drilling performance improvement in offshore batch wells based on rig state classification using machine learning
Document Type
Article
Publication Date
9-1-2020
Abstract
In this study, a novel Artificial Neural Network (ANN) model is developed for rig state classification and its utility as an efficient method is demonstrated for rig crew performance evaluation. The input characteristic vector of ANN is composed of comprehensive logging data. The structure and parameters of ANN are determined according to the output characteristics of rig state that are used to predict in real-time the rig state of the offshore batch wells. Next, the operational time is analyzed, the histograms of operational time components are visualized and the rig crew performance evaluation is conducted. Additionally, the Invisible Lost Time (ILT) is detected and reduced by comparing the operational time and Key Performance Indicators (KPIs) (as designed and set by the operator) to improve the overall drilling performance. The accuracy of developed ANN model is approximately 93%. Finally, the ILT decreases by 45.23% and the overall drilling performance improves by 31.19% through the application of the ANN model for the rig crew performance evaluation.
Publication Source (Journal or Book title)
Journal of Petroleum Science and Engineering
Recommended Citation
Yin, Q., Yang, J., Hou, X., Tyagi, M., Zhou, X., Cao, B., Sun, T., Li, L., & Xu, D. (2020). Drilling performance improvement in offshore batch wells based on rig state classification using machine learning. Journal of Petroleum Science and Engineering, 192 https://doi.org/10.1016/j.petrol.2020.107306