Accurate and Rapid Forecasts for Geologic Carbon Storage via Learning-Based Inversion-Free Prediction

Carbon capture and storage (CCS) is one approach being studied by the U.S. Department of Energy to help mitigate global warming. The process involves capturing CO2 emissions from industrial sources and permanently storing them in deep geologic formations (storage reservoirs). However, CCS projects generally target “green field sites,” where there is often little characterization data and therefore large uncertainty about the petrophysical properties and other geologic attributes of the storage reservoir. Consequently, ensemble-based approaches are often used to forecast multiple realizations prior to CO2 injection to visualize a range of potential outcomes. In addition, monitoring data during injection operations are used to update the pre-injection forecasts and thereby improve agreement between forecasted and observed behavior. Thus, a system for generating accurate, timely forecasts of pressure buildup and CO2 movement and distribution within the storage reservoir and for updating those forecasts via monitoring measurements becomes crucial. This study proposes a learning-based prediction method that can accurately and rapidly forecast spatial distribution of CO2 concentration and pressure with uncertainty quantification without relying on traditional inverse modeling. The machine learning techniques include dimension reduction, multivariate data analysis, and Bayesian learning. The outcome is expected to provide CO2 storage site operators with an effective tool for timely and informative decision making based on limited simulation and monitoring data.

Download Full-text

Delay Prediction of Aircrafts Based on Health Monitoring Data

International Journal of Business Analytics and Intelligence ◽

10.21863/ijbai/2016.4.1.015 ◽

2016 ◽

Vol 4 (1) ◽

Cited By ~ 1

Author(s):

B. A. Dattaram ◽

N. Madhusudanan

Keyword(s):

Machine Learning ◽

Health Monitoring ◽

Monitoring Data ◽

Machine Learning Techniques ◽

Learning Techniques ◽

Flight Delay ◽

Delay Prediction ◽

The Relationship

Flight delay is a major issue faced by airline companies. Delay in the aircraft take off can lead to penalty and extra payment to airport authorities leading to revenue loss. The causes for delays can be weather, traffic queues or component issues. In this paper, we focus on the problem of delays due to component issues in the aircraft. In particular, this paper explores the analysis of aircraft delays based on health monitoring data from the aircraft. This paper analyzes and establishes the relationship between health monitoring data and the delay of the aircrafts using exploratory analytics, stochastic approaches and machine learning techniques.

Download Full-text

Identifying the underlying factors associated with antidepressant drug discontinuation: Content analysis of patients’ drug reviews (Preprint)

10.2196/preprints.23572 ◽

2020 ◽

Author(s):

Mohammad Alarifi ◽

Somaieh Goudarzvand3 ◽

Abdulrahman Jabour ◽

Doreen Foy ◽

Maryam Zolnoori

Keyword(s):

Machine Learning ◽

Antidepressant Drug ◽

Prediction Method ◽

Analytical Framework ◽

Structured Data ◽

Withdrawal Symptoms ◽

Machine Learning Techniques ◽

Drug Discontinuation ◽

Factors Associated ◽

Learning Techniques

BACKGROUND The rate of antidepressant prescriptions is globally increasing. A large portion of patients stop their medications which could lead to many side effects including relapse, and anxiety. OBJECTIVE The aim of this was to develop a drug-continuity prediction model and identify the factors associated with drug-continuity using online patient forums. METHODS We retrieved 982 antidepressant drug reviews from the online patient’s forum AskaPatient.com. We followed the Analytical Framework Method to extract structured data from unstructured data. Using the structured data, we examined the factors associated with antidepressant discontinuity and developed a predictive model using multiple machine learning techniques. RESULTS We tested multiple machine learning techniques which resulted in different performances ranging from accuracy of 65% to 82%. We found that Radom Forest algorithm provides the highest prediction method with 82% Accuracy, 78% Precision, 88.03% Recall, and 84.2% F1-Score. The factors associated with drug discontinuity the most were; withdrawal symptoms, effectiveness-ineffectiveness, perceived-distress-adverse drug reaction, rating, and perceived-distress related to withdrawal symptoms. CONCLUSIONS Although the nature of data available at online forums differ from data collected through surveys, we found that online patients forum can be a valuable source of data for drug-continuity prediction and understanding patients experience. The factors identified through our techniques were consistent with the findings of prior studies that used surveys.

Download Full-text

Assessing the Accuracy of Fault Interpretation using Machine Learning Techniques when Risking Faults for CO2 Storage Site Assessment

Interpretation ◽

10.1190/int-2021-0077.1 ◽

2021 ◽

pp. 1-55

Author(s):

Emma A. H. Michie ◽

Behzad Alaei ◽

Alvar Braathen

Keyword(s):

Deep Learning ◽

New Technologies ◽

Co2 Storage ◽

Fault Reactivation ◽

Machine Learning Techniques ◽

Close Similarity ◽

Storage Site ◽

Site Assessment ◽

Learning Techniques ◽

Fault Interpretation

Generating an accurate model of the subsurface for the purpose of assessing the feasibility of a CO2 storage site is crucial. In particular, how faults are interpreted is likely to influence the predicted capacity and integrity of the reservoir; whether this is through identifying high risk areas along the fault, where fluid is likely to flow across the fault, or by assessing the reactivation potential of the fault with increased pressure, causing fluid to flow up the fault. New technologies allow users to interpret faults effortlessly, and in much quicker time, utilizing methods such as Deep Learning. These Deep Learning techniques use knowledge from Neural Networks to allow end-users to compute areas where faults are likely to occur. Although these new technologies may be attractive due to reduced interpretation time, it is important to understand the inherent uncertainties in their ability to predict accurate fault geometries. Here, we compare Deep Learning fault interpretation versus manual fault interpretation, and can see distinct differences to those faults where significant ambiguity exists due to poor seismic resolution at the fault; we observe an increased irregularity when Deep Learning methods are used over conventional manual interpretation. This can result in significant differences between the resulting analyses, such as fault reactivation potential. Conversely, we observe that well-imaged faults show a close similarity between the resulting fault surfaces when both Deep Learning and manual fault interpretation methods are employed, and hence we also observe a close similarity between any attributes and fault analyses made.

Download Full-text

Assessing the Effect of Velocity Model Accuracy on Microseismic Interpretation at the In Salah Carbon Capture and Storage Site

Energy Procedia ◽

10.1016/j.egypro.2014.11.473 ◽

2014 ◽

Vol 63 ◽

pp. 4385-4393 ◽

Cited By ~ 7

Author(s):

Anna L. Stork ◽

James P. Verdon ◽

J.-Michael Kendall

Keyword(s):

Carbon Capture ◽

Carbon Capture And Storage ◽

Velocity Model ◽

Storage Site ◽

Model Accuracy ◽

And Storage

Download Full-text

A Prediction Method of Solar Power Generator using Machine Learning Techniques

Parallel and Distributed Computing, Applications and Technologies - Communications in Computer and Information Science ◽

10.1007/978-981-13-5907-1_37 ◽

2019 ◽

pp. 345-352

Author(s):

Jungseok Cho ◽

Jeongdoo Lee ◽

Doosan Cho

Keyword(s):

Machine Learning ◽

Solar Power ◽

Prediction Method ◽

Machine Learning Techniques ◽

Power Generator ◽

Learning Techniques

Download Full-text

The Endurance CO2 storage site, Blocks 42/25 and 43/21, UK North Sea

Geological Society London Memoirs ◽

10.1144/m52-2019-47 ◽

2020 ◽

Vol 52 (1) ◽

pp. 163-171 ◽

Cited By ~ 1

Author(s):

Jon G. Gluyas ◽

Usman Bagudu

Keyword(s):

Carbon Capture ◽

Co2 Storage ◽

Carbon Capture And Storage ◽

Lower Triassic ◽

Injection Rate ◽

Government Support ◽

Storage Site ◽

Sandstone Formation ◽

The Uk ◽

And Storage

AbstractThe Endurance, four-way, dip-closed structure in UK Blocks 42/25 and 43/21 occurs over a salt swell diapir and within Triassic and younger strata. The Lower Triassic Bunter Sandstone Formation reservoir within the structure was tested twice for natural gas (in 1970 and 1990) but both wells were dry. The reservoir is both thick and high quality and, as such, an excellent candidate site for subsurface CO2 storage.In 2013 a consortium led by National Grid Carbon drilled an appraisal well on the structure and undertook an injection test ahead of a planned development of Endurance as the first bespoke storage site on the UK Continental Shelf with an expected injection rate of 2.68 × 106 t of dense phase CO2 each year for 20 years. The site was not developed following the UK Government's removal of financial support for carbon capture and storage (CCS) demonstration projects, but it is hoped with the recent March 2020 Budget that government support for CCS may now be back on track.

Download Full-text

Uncertainty in regional estimates of capacity for carbon capture and storage

Solid Earth ◽

10.5194/se-10-1707-2019 ◽

2019 ◽

Vol 10 (5) ◽

pp. 1707-1715 ◽

Cited By ~ 1

Author(s):

Mark Wilkinson ◽

Debbie Polson

Keyword(s):

Carbon Capture ◽

Storage Capacity ◽

Significant Proportion ◽

Carbon Capture And Storage ◽

Saline Aquifers ◽

Storage Site ◽

Limited Data ◽

Wide Range ◽

Regional Estimates ◽

And Storage

Abstract. Carbon capture and storage (CCS) is a potentially important technology for the mitigation of industrial CO2 emissions. However, the majority of the subsurface storage capacity is in saline aquifers, for which there is relatively little information. Published estimates of the potential storage capacity of such formations, based on limited data, often give no indication of the uncertainty, despite there being substantial uncertainty associated with the data used to calculate such estimates. Here, we test the hypothesis that the uncertainty in such estimates is a significant proportion of the estimated storage capacity, and should hence be evaluated as a part of any assessment. Using only publicly available data, a group of 13 experts independently estimated the storage capacity of seven regional saline aquifers. The experts produced a wide range of estimates for each aquifer due to a combination of using different published values for some variables and differences in their judgements of the aquifer properties such as area and thickness. The range of storage estimates produced by the experts shows that there is significant uncertainty in such estimates; in particular, the experts' range does not capture the highest possible capacity estimates. This means that by not accounting for uncertainty, such regional estimates may underestimate the true storage capacity. The result is applicable to single values of storage capacity of regional potential but not to detailed studies of a single storage site.

Download Full-text

USING SEISMIC ATTRIBUTES TO EVALUATE LOWER PALEOZOIC AND PRECAMBRIAN STRUCTURE BENEATH A CARBON CAPTURE AND STORAGE SITE, ILLINOIS BASIN

10.1130/abs/2018am-321364 ◽

2018 ◽

Author(s):

R. William Keach ◽

◽

John H. McBride ◽

Hannes E. Leetaru

Keyword(s):

Carbon Capture ◽

Carbon Capture And Storage ◽

Seismic Attributes ◽

Illinois Basin ◽

Storage Site ◽

Lower Paleozoic ◽

And Storage

Download Full-text

South West Hub Project: appraising a carbon storage resource in Western Australia

The APPEA Journal ◽

10.1071/aj14107 ◽

2015 ◽

Vol 55 (2) ◽

pp. 472

Author(s):

Linda Stalker ◽

Dominique Van Gent ◽

Sandeep Sharma ◽

Martin Burke

Keyword(s):

Carbon Storage ◽

Carbon Capture ◽

Local Community ◽

Co2 Storage ◽

Carbon Capture And Storage ◽

Relevant Information ◽

Seismic Survey ◽

Storage Site ◽

South West ◽

Commercial Scale

The South West Hub Carbon Capture and Storage Project (SWH), managed by the WA Department of Mines and Petroleum (WA DMP), is evaluating the potential for a commercial-scale carbon storage site near major emissions sites in southwest WA. The area under investigation is in the southern Perth Basin, focusing on a 150 km2 area in the shires of Harvey and Waroona. WA DMP is conducting a major feasibility study and collecting pre-competitive data in partnership with the local community. The activities are done in a stage-gate model to obtain relevant information on the potential storage capacity, containment security and injectivity of the geology. Following a smaller 2D seismic survey and the drilling of the Harvey–1 stratigraphic well, a more complex 3D seismic survey was undertaken in February to March, 2014. These activities have confirmed the potential for commercial-scale CO2 storage. A new work package has been initiated with the drilling of three wells (Harvey–2, –3 and –4) underway and plans to drill a fifth well in the next 12 months. The stage-gate approach has been cost-effective, resulting in a carefully planned data acquisition and research program. The approach allows new results, information and potential future activities to be rolled out to stakeholders and the community in the area.

Download Full-text