This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 208689, “Coupling Psychological Factors With Machine Learning To Improve Rig Technical Training,” by Francesco Curina, Elia Abdo, and Ajith Asokan, Drillmec, et al. The paper has not been peer reviewed.
The complete paper explains how machine learning and physiology can be used to improve rig technical training by monitoring the operator’s stress, identifying key operations in which situational awareness is low, and targeting these operations with dedicated exercises. The developed methodology is based on a study of human psychological indicators captured through light biometric devices. These indicators are fed to a machine-learning algorithm that calculates a stress index for the observed operator. The model uses machine vision to identify key physiological parameters and a convolutional neural network to interpret them. Finally, a third algorithm correlates the stress index to specific operations.
In recent years, with the advent of machine learning and accurate biometric sensing devices, many models have been produced to identify stress automatically and instantaneously. Support-vector-machine and random forest models have shown special affinity for these applications. As for biometric sensing devices, they currently are found in most wearables and are available for relatively low prices. Most notably, electrocardiogram (ECG), electroencephalogram, and photoplethysmograph sensors, in addition to thermal cameras, are the most-used types.
In drilling operations, a driller may be aware of emergency procedures but the actual event itself will stress them acutely. In addition, continuous changes introduced to the layout or technology of the human/machine interface (HMI) can be a source of stress even for an experienced driller. Thus, identification of the stressful scenario for a specific driller and training them for that task while using the same HMI as the actual rig would enable the operator to stay alert and responsive during a real emergency on the rig.
Many emergency situations may occur during drilling operations that may lead to high-risk situations. However, none can be considered a higher risk than a well-control situation. Therefore, the focus of this study has been placed upon well-control situations and simulator training.