Domain-specific data quality metrics are key to data analytics efforts in materials science. To fill this need, metrics were developed through the eXtremeMAT project to assess the relative quality of alloy composition, processing, and experimental testing data. Understanding the quality of experimental alloy data allows only the highest quality information to be integrated into analytical models, while low to medium quality data can be reserved for model assessment and validation. Stratification of data based on quality is critical as data is integrated from numerous resources with varying data types and formats. Specific metrics were developed to assign a quality rating for the completeness, usability, accuracy, and standardization of the data. To enable researchers to easily rate a dataset, a data quality rating tool was developed and integrated into EDX.

• eXtremeMAT developed code can simulate the response of an alloy subjected to creep loading conditions for a time period of 10 years in approximately 5 hours. The model has been successfully calibrated for alloy 347H and is being used to produce a database of expected rupture life as a function of stress and temperature.

For more information on NETL’s affordable, durable alloys please go to https://edx.netl.doe.gov/sites/extrememat/.

The SMART Real-Time Forecasting Team is developing a learning-based inversion-free prediction (LIP) framework that can efficiently produce real-time forecasting with uncertainty quantification informed by streaming observation data through parallel forward simulations. The key idea of LIP framework is to circumvent the challenging inverse modeling by precomputing an ensemble of unconstrained forward simulations and then using ML methods to learn the complex relationship between simulated observation and prediction variables. Once the ML model has learned the relationship, it can be used to continually update predictions of future system behavior based on streaming and multiple sources of observation data to enable rapid data assimilation and real-time decision support.

For more information on NETL’s SMART Initiative please go to https://edx.netl.doe.gov/sites/smart/.

NETL’s Offshore Risk Modeling (ORM) suite resulted in a flexible set of custom data, tools, and models that integrate innovative spatio-temporal analytics, ML, big data, and advanced visualization technologies to support DOE’s offshore spill prevention, operational efficiency, and safety goals. Five years of development produced terabytes of new data and seven trademarked or copyrighted tools built into the ORM suite that can be used independently or in combination to support data-driven analytics for offshore systems to improve global energy, environmental, and economic conditions. NETL has demonstrated how the ORM suite can be used to help improve reserves estimates, increase profitability, guide safety and maintenance decisions, forecast risks, and optimize well/facilities designs. These pioneering applications of the ORM suite, and the data science innovations driving them, have garnered national and global attention. This has translated into millions of dollars of funding from DOE-FE and external stakeholders for new projects that apply the ORM suite to address additional energy systems and help inform a range of industry and regulatory decisions. To date, the ORM suite has been adapted to address energy infrastructure, carbon storage, geothermal, rare-earth element, induced seismicity, energy materials, and other oil and gas system needs.

For more information on NETL’s Advanced Offshore Research Portfolio please go to: https://edx.netl.doe.gov/sites/offshore/

The Institute for the Design of Advanced Energy Systems (IDAES) provides revolutionary new capabilities for Process Systems Engineering that exceed existing tools and approaches. After years of development, IDAES has established an extensive modeling library integrated with a suite of advanced ML, uncertainty quantification and optimization techniques that enable process optimization, parameter estimation, advanced process control and multi-scale analysis across various levels of implementation. With its equation-oriented modeling approach, IDAES uniquely supports the process modeling lifecycle, from conceptual design to dynamic optimization and control. IDAES has cultivated an active, growing user community from multiple industries.

• IDAES uses AI and machine learning models as surrogates for complex unit operations or constraints to increase the size and complexity of models that can be optimized within IDAES.
• IDAES is developing AI approaches for improved model and solver performance.

For more information on NETL’s Institute for the Design of Advanced Energy Systems please go to: https://idaes.org/

The Carbon Capture Simulation for Industry Impact (CCSI²) is an open-source computational toolset to accelerate and de-risk technology development and commercialization through first-principles, multi-scale modeling, optimization and uncertainty quantification. CCSI²’s open-source computational toolset provides end users in industry with a comprehensive, integrated suite of scientifically-validated models with uncertainty quantification, optimization, risk analysis, and intelligent decision-making capabilities. CCSI² utilizes a variety of ML methods within its Framework for Quantification of Uncertainty and Surrogates (FOQUS), helping generate optimal experimental designs that maximize the learning from costly laboratory and pilot scale experiments, reducing technical risk. CCSI² also employs ML to accelerate the solution of complex models for the detailed design of novel carbon capture devices and components that utilize advanced manufacturing to enable process intensification.

For more information on NETL’s Carbon Capture Simulation for Industry Impact please go to: https://www.netl.doe.gov/coal/carbon-capture/ccsi2