Rapid and Efficient Waterflood Optimization Using Augmented AI Approach in a Complex Offshore Field

2021 ◽  
Author(s):  
Shawket Ghedan ◽  
Meher Surendra ◽  
Agustin Maqui ◽  
Mahmoud Elwan ◽  
Rami Kansao ◽  
...  
Keyword(s):  
Machine Learning ◽  
Oil Field ◽  
Machine Learning Algorithms ◽  
Data Driven ◽  
Water Flooding ◽  
Fault System ◽  
Fractional Flow ◽  
Operational Decisions ◽  
Water Cut ◽  
Offshore Field

Abstract Waterfloods are amongst the most widely implemented methods for oil field development. Despite their vast implementation, operational bottlenecks such as lack of surveillance and optimization tools to guide fast paced decisions render most of these sub-optimal. This paper presents a novel machine-learning, reduced-physics approach to optimize an exceptionally complex off-shore waterflood in the Gulf of Suez. Leveraging a hybrid data-driven and physics approach, the water flooding scheme in Nezzezat reservoir was optimized to improve reservoir voidage replacement, increase oil production, and reduce water production by identifying potential in wells. As a by-product of the study, a better understanding of the complex fault system was also achieved. Including the geological understanding and its uncertainty is one of the key elements that must be preserved. All geological attributes, along with production rates are used to solve for pressure and inter-well communication. This is later supplemented by machine-learning algorithm to solve for the fractional flow of inter-well connections. Combining the inter-well connectivity and fractional flow, an optimization was performed to reach the best possible conditions for oil gains and water-cut reduction. A global optimization is possible thanks to the low computational demand of this approach, as thousands to millions of realizations must be run to reach the best solution while satisfying all constraints. This is all done in a fraction of the time it takes to run a traditional reservoir simulation. For the present case, the paper will present the underlying physics and data-driven algorithms, along with the blind tests performed to validate the results. In addition to the method's inner workings, the paper will focus more on the results to guide operational decisions. This is inclusive of all the complex constraints of an offshore field, as well as the best reservoir management practices, when reaching optimal production and injection rates for each well. An increase in production was achieved with some reduction in water-cut, while honoring well and platform level limitations. While these represent the gains for a particular month, optimization scenarios can be run weekly or monthly to capture the dynamic nature of the problem and any operational limitations that might arise. The ability to update the models and run optimization scenarios effortlessly allows pro-active operational decisions to maximize the value of the asset. The approach followed in this paper solves for the critical physics of the problem and supplements the remaining with machine learning algorithms. This novel and extremely practical approach facilitate the decision making to operate the field optimally.

scholarly journals Predicting rice blast disease: machine learning versus process-based models

2019 ◽  
Vol 20 (1) ◽  
Author(s):  
David F. Nettleton ◽  
Dimitrios Katsantonis ◽  
Argyris Kalaitzidis ◽  
Natasa Sarafijanovic-Djukic ◽  
Pau Puigdollers ◽  
...  
Keyword(s):  
Machine Learning ◽  
Rice Blast ◽  
Machine Learning Algorithms ◽  
Rice Blast Disease ◽  
Blast Disease ◽  
Data Driven ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Plant Disease Management ◽  
Process Based Models

Abstract Background In this study, we compared four models for predicting rice blast disease, two operational process-based models (Yoshino and Water Accounting Rice Model (WARM)) and two approaches based on machine learning algorithms (M5Rules and Recurrent Neural Networks (RNN)), the former inducing a rule-based model and the latter building a neural network. In situ telemetry is important to obtain quality in-field data for predictive models and this was a key aspect of the RICE-GUARD project on which this study is based. According to the authors, this is the first time process-based and machine learning modelling approaches for supporting plant disease management are compared. Results Results clearly showed that the models succeeded in providing a warning of rice blast onset and presence, thus representing suitable solutions for preventive remedial actions targeting the mitigation of yield losses and the reduction of fungicide use. All methods gave significant “signals” during the “early warning” period, with a similar level of performance. M5Rules and WARM gave the maximum average normalized scores of 0.80 and 0.77, respectively, whereas Yoshino gave the best score for one site (Kalochori 2015). The best average values of r and r2 and %MAE (Mean Absolute Error) for the machine learning models were 0.70, 0.50 and 0.75, respectively and for the process-based models the corresponding values were 0.59, 0.40 and 0.82. Thus it has been found that the ML models are competitive with the process-based models. This result has relevant implications for the operational use of the models, since most of the available studies are limited to the analysis of the relationship between the model outputs and the incidence of rice blast. Results also showed that machine learning methods approximated the performances of two process-based models used for years in operational contexts. Conclusions Process-based and data-driven models can be used to provide early warnings to anticipate rice blast and detect its presence, thus supporting fungicide applications. Data-driven models derived from machine learning methods are a viable alternative to process-based approaches and – in cases when training datasets are available – offer a potentially greater adaptability to new contexts.

scholarly journals Effectiveness of data-driven induction of semantic spaces and traditional classifiers for sarcasm detection

2019 ◽  
Vol 25 (2) ◽  
pp. 257-285 ◽  
Author(s):  
Mattia Antonino Di Gangi ◽  
Giosué Lo Bosco ◽  
Giovanni Pilato
Keyword(s):  
Machine Learning ◽  
Scientific Community ◽  
Semantic Space ◽  
Machine Learning Algorithms ◽  
Data Driven ◽  
Detection Problem ◽  
Everyday Language ◽  
Text Understanding ◽  
The Social ◽  
Semantic Spaces

AbstractIrony and sarcasm are two complex linguistic phenomena that are widely used in everyday language and especially over the social media, but they represent two serious issues for automated text understanding. Many labeled corpora have been extracted from several sources to accomplish this task, and it seems that sarcasm is conveyed in different ways for different domains. Nonetheless, very little work has been done for comparing different methods among the available corpora. Furthermore, usually, each author collects and uses their own datasets to evaluate his own method. In this paper, we show that sarcasm detection can be tackled by applying classical machine-learning algorithms to input texts sub-symbolically represented in a Latent Semantic space. The main consequence is that our studies establish both reference datasets and baselines for the sarcasm detection problem that could serve the scientific community to test newly proposed methods.

Evolving the Materials Genome: How Machine Learning Is Fueling the Next Generation of Materials Discovery

2020 ◽  
Vol 50 (1) ◽  
pp. 1-25 ◽  
Author(s):  
Changwon Suh ◽  
Clyde Fare ◽  
James A. Warren ◽  
Edward O. Pyzer-Knapp
Keyword(s):  
Machine Learning ◽  
Materials Science ◽  
State Of The Art ◽  
Learning Technologies ◽  
Machine Learning Algorithms ◽  
Data Driven ◽  
Computational Learning ◽  
Data Intensive ◽  
Research Activities ◽  
Materials Genome

Machine learning, applied to chemical and materials data, is transforming the field of materials discovery and design, yet significant work is still required to fully take advantage of machine learning algorithms, tools, and methods. Here, we review the accomplishments to date of the community and assess the maturity of state-of-the-art, data-intensive research activities that combine perspectives from materials science and chemistry. We focus on three major themes—learning to see, learning to estimate, and learning to search materials—to show how advanced computational learning technologies are rapidly and successfully used to solve materials and chemistry problems. Additionally, we discuss a clear path toward a future where data-driven approaches to materials discovery and design are standard practice.

scholarly journals Machine Learning-Based Probabilistic Lithofacies Prediction from Conventional Well Logs: A Case from the Umiat Oil Field of Alaska

Energies ◽  
2020 ◽  
Vol 13 (18) ◽  
pp. 4862
Author(s):  
Nilesh Dixit ◽  
Paul McColgan ◽  
Kimberly Kusler
Keyword(s):  
Machine Learning ◽  
Oil Field ◽  
Machine Learning Algorithms ◽  
Self Organizing Maps ◽  
Log Data ◽  
Core Data ◽  
Core Samples ◽  
Resolution Graph ◽  
Reservoir Facies ◽  
Self Organizing

A good understanding of different rock types and their distribution is critical to locate oil and gas accumulations in the subsurface. Traditionally, rock core samples are used to directly determine the exact rock facies and what geological environments might be present. Core samples are often expensive to recover and, therefore, not always available for each well. Wireline logs provide a cheaper alternative to core samples, but they do not distinguish between various rock facies alone. This problem can be overcome by integrating limited core data with largely available wireline log data with machine learning. Here, we presented an application of machine learning in rock facies predictions based on limited core data from the Umiat Oil Field of Alaska. First, we identified five sandstone reservoir facies within the Lower Grandstand Member using core samples and mineralogical data available for the Umiat 18 well. Next, we applied machine learning algorithms (ascendant hierarchical clustering, self-organizing maps, artificial neural network, and multi-resolution graph-based clustering) to available wireline log data to build our models trained with core-driven information. We found that self-organizing maps provided the best result among other techniques for facies predictions. We used the best self-organizing maps scheme for predicting similar reservoir facies in nearby uncored wells—Umiat 23H and SeaBee-1. We validated our facies prediction results for these wells with observed seismic data.

Oil-Rate Forecast by Inferring Fractional-Flow Models From Field Data With Koval Method Combined With the Capacitance/Resistance Model

2015 ◽  
Vol 18 (04) ◽  
pp. 534-553 ◽  
Author(s):  
Fei Cao ◽  
Haishan Luo ◽  
Larry W. Lake
Keyword(s):  
Field Data ◽  
Oil Production ◽  
Data Driven ◽  
Analytical Models ◽  
Fractional Flow ◽  
Physical Behavior ◽  
Flow Models ◽  
Drainage Volume ◽  
Resistance Model ◽  
Water Cut

Summary Many empirical and analytical models were developed to forecast oil production. Empirical models (including data-driven models) can, for example, find correlations between oil cut and production, but they lack explicit knowledge of the physical behavior. Classic analytical models are loyal to reservoir physics. Nevertheless, they often require estimation of water saturation as a function of time, which is difficult to obtain for multiwell reservoirs. It is desirable to combine advantages of both empirical and analytical models and develop a physical-model-based method that uses field data to infer oil rate. In this paper, we propose to infer fractional-flow models from field data by use of the Koval (1963) theory. We inversely solved the Koval fractional-flow equation to obtain a relationship between water cut and dimensionless time. By history matching field water-cut data, two model parameters, the Koval factor and the producer-drainage volume, are estimated. Nevertheless, it is challenging to use the Koval approach as a predictive model directly because the injection contribution into each producer in a future-time horizon must be evaluated first. To address this issue, we combine the Koval approach with the capacitance/resistance model (CRM), which characterizes the injector/producer connectivities and response time. The material balance of fluids is established in a producer-based drainage volume to consider the contributions from nearby injectors and the time lag in production caused by reservoir/fluids compressibility. A regression approach is simultaneously advanced to minimize the model error. Because of robustly integrating the reservoir physical behavior and the data-driven approach, the combination of the Koval theory and the CRM can result in a synergy that leads to accurate oil-rate predictions. We validated this integrated method in synthetic homogeneous and heterogeneous reservoirs to test its reliability, and further applied it to a field case in western Venezuela. Case studies demonstrate that one can use this integrated model as a real-time tool to characterize interwell connection and to predict future oil production accurately.

scholarly journals Machine learning to explain printability induced by rheology additives

2021 ◽  
Author(s):  
Ali Nadernezhad ◽  
Jürgen Groll
Keyword(s):  
Machine Learning ◽  
Decision Making ◽  
Machine Learning Algorithms ◽  
Data Driven ◽  
Acid Solutions ◽  
Continuous Growth ◽  
Multiple Systems ◽  
Molecular Weights ◽  
Underlying Basis

With the continuous growth of extrusion bioprinting techniques, ink formulations based on rheology modifiers are becoming increasingly popular, as they enable 3D printing of non-printable biologically-favored materials. However, benchmarking and characterization of such systems are inherently complicated due to the variety of rheology modifiers and differences in mechanisms of inducing printability. This study tries to explain induced printability in formulations by incorporating machine learning algorithms that describe the underlying basis for decision-making in classifying a printable formulation. For this purpose, a library of rheological data and printability scores for 180 different formulations of hyaluronic acid solutions with varying molecular weights and concentrations and three rheology modifiers were produced. A feature screening methodology was applied to collect and separate the impactful features, which consisted of physically interpretable and easily measurable properties of formulations. In the final step, all relevant features influencing the model’s output were analyzed by advanced yet explainable statistical methods. The outcome provides a guideline for designing new formulations based on data-driven correlations from multiple systems.

Hybrid machine learning algorithms to enhance lost-circulation prediction and management in the Marun oil field

2021 ◽  
Vol 198 ◽  
pp. 108125
Author(s):  
Mohammad Sabah ◽  
Mohammad Mehrad ◽  
Seyed Babak Ashrafi ◽  
David A. Wood ◽  
Shadi Fathi
Keyword(s):  
Machine Learning ◽  
Learning Algorithms ◽  
Oil Field ◽  
Machine Learning Algorithms ◽  
Lost Circulation ◽  
Hybrid Machine

scholarly journals Predictive Pattern Classification Can Distinguish Gender Identity Subtypes from Behavior and Brain Imaging

2020 ◽  
Vol 30 (5) ◽  
pp. 2755-2765 ◽  
Author(s):  
Benjamin Clemens ◽  
Birgit Derntl ◽  
Elke Smith ◽  
Jessica Junger ◽  
Josef Neulen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Functional Connectivity ◽  
Gender Identity ◽  
Machine Learning Algorithms ◽  
Data Driven ◽  
Subjective Perception ◽  
Questionnaire Data ◽  
Resting State Functional Connectivity ◽  
Gender Dimorphism ◽  
Transgender Persons

Abstract The exact neurobiological underpinnings of gender identity (i.e., the subjective perception of oneself belonging to a certain gender) still remain unknown. Combining both resting-state functional connectivity and behavioral data, we examined gender identity in cisgender and transgender persons using a data-driven machine learning strategy. Intrinsic functional connectivity and questionnaire data were obtained from cisgender (men/women) and transgender (trans men/trans women) individuals. Machine learning algorithms reliably detected gender identity with high prediction accuracy in each of the four groups based on connectivity signatures alone. The four normative gender groups were classified with accuracies ranging from 48% to 62% (exceeding chance level at 25%). These connectivity-based classification accuracies exceeded those obtained from a widely established behavioral instrument for gender identity. Using canonical correlation analyses, functional brain measurements and questionnaire data were then integrated to delineate nine canonical vectors (i.e., brain-gender axes), providing a multilevel window into the conventional sex dichotomy. Our dimensional gender perspective captures four distinguishable brain phenotypes for gender identity, advocating a biologically grounded reconceptualization of gender dimorphism. We hope to pave the way towards objective, data-driven diagnostic markers for gender identity and transgender, taking into account neurobiological and behavioral differences in an integrative modeling approach.

Sign in / Sign up

Export Citation Format

Share Document