Semi-Supervised Approach for Early Stuck Sign Detection in Drilling Operations

TL;DR

This paper presents a semi-supervised real-time method for early detection of stuck pipe in drilling operations by analyzing deviations in drilling data using auto-encoder models, outperforming previous supervised approaches.

Contribution

It introduces a semi-supervised approach utilizing auto-encoders for early stuck pipe detection, adapting to varying geological conditions and demonstrating improved performance over prior methods.

Findings

Eight stuck incidents showed large reconstruction errors

The approach outperforms previous supervised methods

Model robustness varies with featured parameters

Abstract

A real-time stuck pipe prediction methodology is proposed in this paper. We assume early signs of stuck pipe to be apparent when the drilling data behavior deviates from that from normal drilling operations. The definition of normalcy changes with drill string configuration or geological conditions. Here, a depth-domain data representation is adopted to capture the localized normal behavior. Several models, based on auto-encoder and variational auto-encoders, are trained on regular drilling data extracted from actual drilling data. When the trained model is applied to data sets before stuck incidents, eight incidents showed large reconstruction errors. These results suggest better performance than the previously reported supervised approach. Inter-comparison of various models reveals the robustness of our approach. The model performance depends on the featured parameter suggesting the need for multiple models in actual operation.