Camera-Based Edge Analytics for Drilling Optimization

Gooneratne, Chinthaka P.; Magana-Mora, Arturo; Affleck, Michael; Otalvora, William Contreras; Zhan, Guo‐Dong; Moellendick, Timothy

doi:10.1109/edge50951.2020.00024

Cited by 6 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Marketing domain has two papers, one on electronic scent diffusers (P2 [16]), and another one on market basket analysis (P12 [1]). Regarding industry domain, we have found one paper on drilling automation in oil and gas industry (P7 [12]) and one paper on IC test in semiconductor manufacturing (P21 [14]).…”

Section: Applications and Domains Of Edge Analyticsmentioning

confidence: 99%

Applications of edge analytics: a systematic review

Andročec

2023

Acta Universitatis Sapientiae, Informatica

View full text Add to dashboard Cite

With the development and expansion of the Internet of Things, computing at the edge is becoming increasingly important, especially for applications where real-time response is important. In this paper, we made a systematic review of the literature on analytics at the edge. We extracted data from 40 selected primary relevant studies from the complete set of 419 papers retrieved from scientific databases. In our analysis of the full text of every primary study we investigated: temporal distribution of primary studies, publication types, domain and application areas of the primary papers, used machine learning and deep learning methods. We also elaborated on the main themes of the primary studies and recommended some possible interesting future research possibilities.

show abstract

Section: Applications and Domains Of Edge Analyticsmentioning

confidence: 99%

Applications of edge analytics: a systematic review

Andročec

2023

Acta Universitatis Sapientiae, Informatica

View full text Add to dashboard Cite

show abstract

“…These sources or sensors could also belong to different drilling rig contractors and service providers who collaborate to complete a variety of tasks and operations. As a result, drilling databases host a collection of records stemming from multiple sources, which results in asynchronous data (the clock in every microcontroller is slightly different) and potentially inconsistent sensor sampling frequency [28,29]. Whereas these problems could be resolved or alleviated by transmitting a common pulse to sensors to collect records simultaneously, this is not a standard practice in drilling operations.…”

Section: B Literature Reviewmentioning

confidence: 99%

Deep Learning and Time-Series Analysis for the Early Detection of Lost Circulation Incidents During Drilling Operations

et al. 2021

Self Cite

View full text Add to dashboard Cite

Drilling operations consist of breaking the rock to deepen a wellbore for oil or gas extraction. A drilling fluid, circulating from the surface through the drill pipe and from the annulus to the surface, is used to remove rock cuttings and maintain hydrostatic pressure. Drilling fluid lost circulation incidents (LCIs) are major sources of non-productive time (NPT) in drilling operations. These incidents occur due to preexisting natural fractures (vugs, caverns, etc.) and/or drilling-induced hydraulic fractures. The initiation of an LCI could lead to other hazardous drilling phenomena, such as formation influx or kick/blowout, stuck pipe incidents, among others. LCIs are typically monitored at the rig site by observing drilling fluid levels in the fluid tanks. This manual process incurs missing the occurrence or late detection of LCIs. Machine learning (ML) and deep learning (DL) classification algorithms are powerful in processing time-series data and achieving early detection of such temporal phenomena. In this study, we performed a large-scale analysis of the surface drilling and rheology data obtained from historical wells with LCIs. This analysis includes primary and secondary preprocessing steps including, aggressive sampling, feature engineering, and window normalization to derive generalizable DL models for real-time operations. Focal loss was utilized to account for data class imbalance and train robust and generalizable models. The results obtained from different ML/DL algorithms showed that one-dimensional convolutional neural network models resulted in the best performance with state-of-the-art precision, recall, and F1 scores of 87.34%, 73.40%, and 79.77%, respectively, on unseen test drilling data.

show abstract

“…Moreover, the down-sampling of the data collected at the rigs represents a substantial loss of information available for the models, as the trends may be wholly lost due to the low transmission frequency of ~0.2-1 Hz. For instance, erratic torque is an important sign for detecting stuck drillstring events, however, with a frequency of 0.2Hz, it may be difficult to detect it even by sophisticated non-linear ML models [23,24,62].…”

Section: Internet Of Things In the Drilling Ecosystemmentioning

confidence: 99%

Well Control Space Out: A Deep-Learning Approach for the Optimization of Drilling Safety Operations

Magana-Mora

Affleck²,

Ibrahim

et al. 2021

IEEE Access

Self Cite

View full text Add to dashboard Cite

As drilling of new oil and gas wells increase to meet energy demands, it is essential to optimize processes to ensure the health and safety of the crew as well as the protection of the environment. Drilling operations represent a dynamic and challenging environment with natural and mechanical factors that need to be closely managed. Well control refers to the technique employed while drilling for balancing the hydrostatic and formation pressures to prevent the influx of water, gas, or hydrocarbons that would ultimately result in an uncontrolled flow to the surface. In the event of a well control incident, the crew must take proper and prompt actions to mitigate the risks and shut-in the well. In this study, we introduce the Well Control Space Out technology, an internet-of-things (IoT) environment that couples cameras and an edge server to implement stateof-the-art deep-learning models for the real-time processing of video images recording the drillstring. The computational models automatically perform object detection to keep track of key drilling rig components. The results from the video analysis are displayed on a dashboard describing the state and steps to follow in a well control incident without the need for any time-consuming, manual calculations. The internet-of-things edge foundation laid in drilling can be seamlessly expanded to other upstream sectors, where time-sensitive, critical decisions can be made in real-time, in the field, closer to operations. Finally, this technology can be seamlessly integrated with the current technologies to develop an automated closed-loop control system.INDEX TERMS Automation, deep-learning, computer vision, edge computing, internet-of-things, oil and gas drilling, well control.

show abstract

Camera-Based Edge Analytics for Drilling Optimization

Cited by 6 publications

References 26 publications

Applications of edge analytics: a systematic review

Applications of edge analytics: a systematic review

Deep Learning and Time-Series Analysis for the Early Detection of Lost Circulation Incidents During Drilling Operations

Well Control Space Out: A Deep-Learning Approach for the Optimization of Drilling Safety Operations

Contact Info

Product

Resources

About