An official website of the United States government
This study introduces an innovative physics-informed machine learning approach to predict the cumulative oil production, CO2 stored, and CO2 adsorbed during CO2 injection. The developed model can be utilized to maximize both oil recovery and CO2 storage in naturally fractured reservoirs (NDR). While existing research has largely focused on maximizing oil recovery and net present value (NPV) through CO2 injection, our methodology integrates comprehensive physics data to fine-tune gas injection processes. We consider a wide span of reservoir characteristics in five-spot injection pattern. We developed dual-permeability, naturally fractured numerical compositional models incorporating sorption to simulate flow mechanisms during injection accurately. Using Langmuir's adsorption dynamics in our simulation integrates well designs and injection strategies. Data from these simulations were trained using Artificial Neural Networks (ANN) to create a robust production prediction proxy model for different gas injection designs. The ANN model demonstrates high accuracy, reflecting its potential for optimizing oil production and carbon storage. The developed model facilitated the screening and optimization process for the considered system as it provides the optimized design in a few seconds compared to days. This study not only highlights the cost-efficiency of using machine learning algorithms in optimization processes but also deepens our understanding of their potential applications in enhanced oil recovery (EOR) and carbon capture and storage (CCUS). The developed models can be used to provide new insights into the interactions between fluid compositions and the complex physical phenomena of sorption, underscoring the transformative potential of integrating advanced machine learning with traditional petroleum engineering techniques.
more »
« less
Predicting the formation of fractionally doped perovskite oxides by a function-confined machine learning method
Abstract Fractionally doped perovskites oxides (FDPOs) have demonstrated ubiquitous applications such as energy conversion, storage and harvesting, catalysis, sensor, superconductor, ferroelectric, piezoelectric, magnetic, and luminescence. Hence, an accurate, cost-effective, and easy-to-use methodology to discover new compositions is much needed. Here, we developed a function-confined machine learning methodology to discover new FDPOs with high prediction accuracy from limited experimental data. By focusing on a specific application, namely solar thermochemical hydrogen production, we collected 632 training data and defined 21 desirable features. Our gradient boosting classifier model achieved a high prediction accuracy of 95.4% and a high F1 score of 0.921. Furthermore, when verified on additional 36 experimental data from existing literature, the model showed a prediction accuracy of 94.4%. With the help of this machine learning approach, we identified and synthesized 11 new FDPO compositions, 7 of which are relevant for solar thermochemical hydrogen production. We believe this confined machine learning methodology can be used to discover, from limited data, FDPOs with other specific application purposes.
more »
« less
- Award ID(s):
- 2025541
- PAR ID:
- 10356865
- Date Published:
- Journal Name:
- Communications Materials
- Volume:
- 3
- Issue:
- 1
- ISSN:
- 2662-4443
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Solar energetic particle (SEP) events, originating from solar flares and Coronal Mass Ejections, present significant hazards to space exploration and technology on Earth. Accurate prediction of these high‐energy events is essential for safeguarding astronauts, spacecraft, and electronic systems. In this study, we conduct an in‐depth investigation into the application of multimodal data fusion techniques for the prediction of high‐energy SEP events, particularly ∼100 MeV events. Our research utilizes six machine learning (ML) models, each finely tuned for time series analysis, including Univariate Time Series (UTS), Image‐based model (Image), Univariate Feature Concatenation (UFC), Univariate Deep Concatenation (UDC), Univariate Deep Merge (UDM), and Univariate Score Concatenation (USC). By combining time series proton flux data with solar X‐ray images, we exploit complementary insights into the underlying solar phenomena responsible for SEP events. Rigorous evaluation metrics, including accuracy, F1‐score, and other established measures, are applied, along withK‐fold cross‐validation, to ensure the robustness and generalization of our models. Additionally, we explore the influence of observation window sizes on classification accuracy.more » « less
-
This paper describes a generalizable framework for creating context-aware wall-time prediction models for HPC applications. This framework: (a) cost-effectively generates comprehensive application-specific training data, (b) provides an application-independent machine learning pipeline that trains different regression models over the training datasets, and (c) establishes context-aware selection criteria for model selection. We explain how most of the training data can be generated on commodity or contention-free cyberinfrastructure and how the predictive models can be scaled to the production environment with the help of a limited number of resource-intensive generated runs (we show almost seven-fold cost reductions along with better performance). Our machine learning pipeline does feature transformation, and dimensionality reduction, then reduces sampling bias induced by data imbalance. Our context-aware model selection algorithm chooses the most appropriate regression model for a given target application that reduces the number of underpredictions while minimizing overestimation errors. Index Terms—AI4CI, Data Science Workflow, Custom ML Models, HPC, Data Generation, Scheduling, Resource Estimationsmore » « less
-
Electronic materials that exhibit phase transitions between metastable states (e.g., metal-insulator transition materials with abrupt electrical resistivity transformations) are challenging to decode. For these materials, conventional machine learning methods display limited predictive capability due to data scarcity and the absence of features that impede model training. In this article, we demonstrate a discovery strategy based on multi-objective Bayesian optimization to directly circumvent these bottlenecks by utilizing latent variable Gaussian processes combined with high-fidelity electronic structure calculations for validation in the chalcogenide lacunar spinel family. We directly and simultaneously learn phase stability and bandgap tunability from chemical composition alone to efficiently discover all superior compositions on the design Pareto front. Previously unidentified electronic transitions also emerge from our featureless adaptive optimization engine. Our methodology readily generalizes to optimization of multiple properties, enabling co-design of complex multifunctional materials, especially where prior data is sparse.more » « less
-
Non-stoichiometric perovskite oxides have been studied as a new family of redox oxides for solar thermochemical hydrogen (STCH) production owing to their favourable thermodynamic properties. However, conventional perovskite oxides suffer from limited phase stability and kinetic properties, and poor cyclability. Here, we report a strategy of introducing A-site multi-principal-component mixing to develop a high-entropy perovskite oxide, (La1/6Pr1/6Nd1/6Gd1/6Sr1/6Ba1/6)MnO3 (LPNGSB_Mn), which shows desirable thermodynamic and kinetics properties as well as excellent phase stability and cycling durability. LPNGSB_Mn exhibits enhanced hydrogen production (∼77.5 mmol/mol-oxide) compared to (La2/3Sr1/3)MnO3 (∼53.5 mmol / mol-oxide) in a short 1 hour redox duration and high STCH and phase stability for 50 cycles. LPNGSB_Mn possesses a moderate enthalpy of reduction (252.51–296.32 kJ / mol-oxide), a high entropy of reduction (126.95–168.85 J / mol-oxide), and fast surface oxygen exchange kinetics. All A-site cations do not show observable valence changes during the reduction and oxidation processes. This research preliminarily explores the use of one A-site high-entropy perovskite oxide for STCH.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1038/s43246-022-00269-9
