Search NASA⌕ Search

SEARCH · Search NASA

Results for “data reduction methods”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 649 records · Page 36
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Promoting the regulatory acceptance of combined ion and neutron irradiation for material degradation in nuclear reactors

The Advanced Materials and Manufacturing Technologies (AMMT) program within the Department of Energy (DOE) Office of Nuclear Energy has developed its current recommendation for promoting the use of combined ion irradiation and neutron irradiation for the accelerated qualification of materials to be deployed in nuclear reactors. This plan is intended to provide a collaborative path forward that can be adopted by academia, national laboratories, and industry, and has been developed with input from the regulatory research arm of the U.S. Nuclear Regulatory Commission (NRC). To deploy new materials or materials manufactured with new technologies, such as additive manufacturing, materials must be evaluated for reactor-induced degradation from the combination of harsh temperatures, corrosive environments, and radiation fields. However, rapid deployment of materials necessitates accelerated testing methods rather than relying on years of neutron irradiation in a material test reactor. Ion irradiation has demonstrated success in reproducing material microstructure and select property evolution resulting from neutron irradiation with three to four orders of magnitude reduction in time and cost, making it an ideal candidate for accelerated irradiation testing. This presentation provides context governing both the scientific and regulatory aspects of the proposed goal. The discussion is aimed at a broad audience including researchers from industry, national laboratories, and academia. The recommended path forward is presented as a conceptual framework of specific steps. In brief, the strategy entails developing an integrated ion and neutron irradiation test plan for the material property of interest based on the fundamental tenet of the linkage of microstructure and properties in materials. Physics-based modeling interprets ion irradiation data and predicts neutron irradiation microstructure and properties with uncertainty bounds. The first round of testing is sufficient for an initial licensing application using a risk-informed approach, while a minimum required neutron irradiation test plan reduces cost and time requirements. A surveillance program with witness specimens in-reactor provides additional data over time to improve model predictions to higher damage levels and further reduce uncertainty bounds, which can be used for license extensions or longer lifetimes in new license applications.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2022

In situ feature analysis for large-scale multiphase flow simulations

The study of multiphase flow is essential for designing chemical reactors such as fluidized bed reactors (FBR), as a detailed understanding of hydrodynamics is critical for optimizing reactor performance and stability. An FBR allows scientists to conduct different types of chemical reactions involving multiphase materials, especially interaction between gas and solids. During such complex chemical processes, the formation of void regions in the reactor, generally termed as bubbles, is an important phenomenon. The study of these bubbles has a deep implication in predicting the reactor’s overall efficiency. But physical experiments needed to understand bubble dynamics are costly and non-trivial due to the technical difficulties involved and harsh working conditions of the reactors. Therefore, to study such chemical processes and bubble dynamics, a state-of-the-art computational simulation MFIX-Exa is being developed. Despite the proven accuracy of MFIX-Exa in modeling bubbling phenomena, the large-scale output data prohibits the use of traditional post hoc analysis capabilities in both storage and I/O time. Herein, to address these issues and allow the application scientists to explore the bubble dynamics in an efficient and timely manner, we have developed an end-to-end analytics pipeline that enables in situ detection of bubbles, followed by a flexible post hoc visual exploration methodology of bubble dynamics. The proposed method enables interactive analysis of bubbles, along with quantification of several bubble characteristics, enabling experts to understand the bubble interactions in detail. Positive feedback from the experts has indicated the efficacy of the proposed approach for exploring bubble dynamics in very-large-scale multiphase flow simulations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

National Mandates Won’t Save Us!: How to Design Energy Efficiency Policies that Address Institutional Barriers to Change

There are many national policies aimed at driving an increase in energy efficiency (EE) implementation. But how effective are they? Over the past 20 years, the U.S. government has established five legal authorities mandating federal agencies to prioritize energy-efficient products when purchasing. As the largest single buyer of energy-consuming products, energy-efficient purchasing across the federal sector would result in huge energy savings and emissions reductions. Despite this, our research has found that only ~55% of federal purchases currently meet existing requirements. If national mandates are not enough to ensure that federal agencies are buying efficient products, what is? To answer this question, we surveyed 161 procurement and sustainability staff from 26 different federal agencies. Findings revealed that several institutional factors (i.e., the roles, rules, and tools within an organization) play a key role in determining how likely federal buyers are to prioritize EE. Ensuring that federal agencies comply with existing EE requirements is foundational to achieving a clean energy future. This will require more than national mandates, but also the design of policies that successfully identify and address the institutional factors that must be changed in order to increase EE implementation. This paper presents an overview of the data collection and analysis methods for our study, as well as key survey findings and the insights they offer for overcoming institutional barriers to compliance with existing EE requirements. We conclude by discussing how these strategies can be broadly applied to improve the design of EE policies in the U.S. and beyond.

Morabito, Molly↗
DOE OSTI2025

Evaluating the factors influencing accuracy, interpretability, and reproducibility in the use of machine learning classifiers in biology to enable standardization

The complexity and variability of biological data has promoted the increased use of machine learning methods to understand processes and predict outcomes. These same features complicate reliable, reproducible, interpretable, and responsible use of such methods, resulting in questionable relevance of the derived. outcomes. Here we systematically explore challenges associated with applying machine learning to predict and understand biological processes using a well- characterized in vitro experimental system. We evaluated factors that vary while applying machine learning classifers: (1) type of biochemical signature (transcripts vs. proteins), (2) data curation methods (pre- and post-processing), and (3) choice of machine learning classifier. Using accuracy, generalizability, interpretability, and reproducibility as metrics, we found that the above factors significantly mod- ulate outcomes even within a simple model system. Our results caution against the unregulated use of machine learning methods in the biological sciences, and strongly advocate the need for data standards and validation tool-kits for such studies.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Comparative techno-economic and environmental analysis of using nuclear energy and fossil energy with carbon sequestration in U.S. Gulf Coast petroleum refineries

U.S. Gulf Coast refineries have processing capacity of approximately 9.4 million barrels of crude oil per day—about 50% of U.S. national capacity—while consuming natural gas, electricity, and hydrogen, resulting in approximately 100 million metric tonnes (MMT) of onsite CO₂ emissions in 2022. This study evaluates two alternative refinery energy supply pathways: nuclear energy (NE) and fossil energy-derived hydrogen with carbon capture and sequestration (FE-CCS; autothermal reforming), intermittent renewable sources are not considered as they cannot reliably meet continuous industrial heat and hydrogen demands. Using publicly available data for refineries, we conducted a bottom-up, facility-level assessment to estimate energy use by type, well-to-gate refinery emissions, and associated costs of integration. Compared to current refinery operations using natural gas energy supply and conventional hydrogen production (steam methane reforming without sequestration), NE and FE-CCS could produce, respectively, average emissions reductions of 37% and 42%, total abatement costs of $\$$52–$\$$221 and $\$$211–$\$$612/MT CO₂, and additional costs of about $\$$0.2–$\$$6 and $\$$3–$\$$11 per barrel of crude, respectively. Our analysis indicates that 25 out of 27 refineries have lower total additional cost ($\$$/bbl.) for the NE scenario than the FE-CCS scenario, making a strong case for NE integration with petroleum refineries. This work's main contribution is a detailed bottom-up refinery-level analysis method that can be utilized by the worldwide refining industry and stakeholders as they assess different technological options, along with their costs and environmental impacts for a specific refinery operation.

Carbon capture and sequestration↗
DOE OSTI2020

Integrated hydrogeophysical modelling and data assimilation for geoelectrical leak detection

Time-lapse electrical resistivity tomography (ERT) measurements provide indirect observations of hydrological processes in the Earth's shallow subsurface at high spatial and temporal resolution. ERT has been used in the past decades to detect leaks and monitor the evolution of associated contaminant plumes. Specifically, inverted resistivity images allow visualization of the dynamic changes in the structure of the plume. However, existing methods do not allow the direct estimation of leak parameters (e.g. leak rate, location, etc.) and their uncertainties. We propose an ensemble-based data assimilation framework that evaluates proposed hydrological models against observed time-lapse ERT measurements without directly inverting for the resistivities. Each proposed hydrological model is run through the parallel coupled hydro-geophysical simulation code PFLOTRAN-E4D to obtain simulated ERT measurements. The ensemble of model proposals is then updated using an iterative ensemble smoother. In this paper, we demonstrate the proposed framework on synthetic and field ERT data from controlled tracer injection experiments. Our results show that the approach allows joint identification of contaminant source location, initial release time, and solute loading from the cross-borehole time-lapse ERT data, alongside with an assessment of uncertainties in these estimates. We demonstrate a reduction in site-wide uncertainty by comparing the prior and posterior plume mass discharges at a selected image plane. This framework is particularly attractive to sites that have previously undergone extensive geological investigation (e.g., nuclear sites). It is well suited to complement ERT imaging and we discuss practical issues in its application to field problems.

58 GEOSCIENCES↗
DOE OSTI2025

HIGH-FIDELITY SIMULATION OF SOOT FORMATION AND THERMAL RADIATION IN A LABORATORY-SCALE RICH-QUENCH-LEAN BURNER

High-fidelity simulations of a swirl-stabilized turbulent spray flame in a laboratory-scale aero-combustor have been performed to evaluate the predictability of state-of-the-art models in capturing soot formation. The simulations employ a complex chemical mechanism developed for Jet-A with PAH chemistry, coupled with the Hybrid Method of Moments (HMOM) soot model, and a Lagrangian dilute spray model for the fuel injection. Two simulations are performed to compare the results when thermal radiation is neglected or included in the solution with a mean spectral model. Modeling closures for the soot differential diffusion effects in mixture fraction space, as well as turbulence-radiation interaction are also evaluated using the data generated by the simulations. Given the degree of complexity of the simulation, the results showed good agreement with experimental measurements of the spatial distribution of the soot volume fraction ensemble average. A closer agreement with the experiment is observed when thermal radiation is included in the solution. Thermal radiation is observed to reduce the flame temperature and increase the flame intermittency, denoted by the increase in the temperature standard deviation in mixture fraction space. The reduction in temperature also leads to a reduction in PAH production and soot volume fraction. Turbulence is observed to have different effects on radiative emission depending on the mixture fraction. Turbulent scalar fluctuations significantly enhance radiative emission in fuel lean mixtures and can also play a role for fuel rich conditions. The statistical description of the turbulence-radiation interaction, previously proposed in the literature, was observed to correctly reproduce the high-fidelity results. Model coefficients were provided for swirl-stabilized flames. The soot differential diffusion model, previously proposed in the literature, based on the residual between the exact term and its model approximation, was also evaluated. The residual correction term further improved the agreement with exact differential diffusion term evaluated with the high-fidelity simulation data in mixture fraction space. The results suggest that the effective turbulent Lewis number can be equal to unity in simulations of turbulent non-premixed recirculating flames.

Soriano, Bruno [Sandia National Laboratories (SNL)↗
DOE OSTI2022

Induced Superconducting Pairing in Integer Quantum Hall Edge States

Indium arsenide (InAs) near surface quantum wells (QWs) are promising for the fabrication of semiconductor–superconductor heterostructures given that they allow for a strong hybridization between the two-dimensional states in the quantum well and the ones in the superconductor. In this work, we present results for InAs QWs in the quantum Hall regime placed in proximity of superconducting NbTiN. We observe a negative downstream resistance with a corresponding reduction of Hall (upstream) resistance, consistent with a very high Andreev conversion. Further, we analyze the experimental data using the Landauer-Büttiker formalism, generalized to allow for Andreev reflection processes. We attribute the high efficiency of Andreev conversion in our devices to the large transparency of the InAs/NbTiN interface and the consequent strong hybridization of the QH edge modes with the states in the superconductor.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2025

Reduced Order Modeling conditioned on monitored features for response and error bounds estimation in engineered systems

Reduced Order Models (ROMs) form essential tools across engineering domains by virtue of their function as surrogates for computationally intensive digital twinning simulators. Although purely data-driven methods are available for ROM construction, schemes that allow to retain a portion of the physics tend to enhance the interpretability and generalization of ROMs. However, physics-based techniques can adversely scale when dealing with nonlinear systems that feature parametric dependencies. This study introduces a generative physics-based ROM that is suited for nonlinear systems with parametric dependencies and is additionally able to provide numerical error bounds associated with the respective estimates. A main contribution of this work is the conditioning of these parametric ROMs to features that can be derived from monitoring measurements, feasibly in an online fashion. This is contrary to most existing ROM schemes, which remain restricted to the prescription of the physics-based, and usually a priori unknown, system parameters. Our work utilizes conditional Variational Autoencoders to continuously map the required reduction bases to a feature vector extracted from limited output measurements, while additionally allowing for a probabilistic assessment of the ROM-estimated Quantities of Interest. An auxiliary task using a neural network-based parametrization of suitable probability distributions is introduced to re-establish the link with physical model parameters. We verify the proposed scheme on a series of simulated case studies incorporating effects of geometric and material nonlinearity under parametric dependencies related to system properties and input load characteristics.

Conditional VAEs↗
DOE OSTI2021

Potential Infrastructure Cost Savings at Hybrid Wind Plus Solar PV Plants

This report documents the creation of NREL's Balance of System model for Hybrid Renewable Energy Systems (HybridBOSSE). Balance of System (BOS) costs account for approximately 30% of the capital expenditure required to install a land-based wind plant, and as much as 40% of the cost of installing a solar PV plant. Hybrid Renewable Energy systems present an opportunity to combine balance of system components to achieve cost savings, and HybridBOSSE aims to provide a component-level model to determine the associated BOS costs in a hybrid renewable energy system. Prior BOS models for hybrid systems (at NREL and elsewhere) have relied on empirical curve fits of legacy industry data, limiting their utility and their predictive ability for new technology combinations and configurations, such as those found in Hybrid Plants. Hybrid plants at the utility scale have considerably more interest in the last number of years, with ever-growing capacities in the interconnection queues, and many large developers pivoting to only consider hybrid plants going forward. As these plants become a commercial reality, the need for a detailed analysis of costs in a hybrid plant to fully exploit the available cost savings grows. The tool developed and documented in this work, HybridBOSSE, provides an open-source, python-based method of assessing the costs in hybrid plants. This analysis capability unlocks greater certainty in hybrid plant costs, as well as the ability to examine scenarios of hybridization and glean combinations that offer the greatest flexibility and cost reduction.

14 SOLAR ENERGY↗
DOE OSTI2020

Review of heat transfer enhancement techniques for single phase flows

The thermal energy exchange between a flowing fluid and its confining channel is a ubiquitous process in modern society. To enhance the fluid-to-wall or wall-to-fluid heat transfer, several techniques have been developed to maximize the contact area between the fluid and the inner wall and/or disrupt the flow to enhance circulation or induce turbulence. Deployment of channels having features capable of enhancing heat transfer enables the reduction of heat exchanger size while maintaining performance. Reduction in equipment size is critical due to the ability to minimize the required volume of costly working fluids and to mitigate potential safety concerns associated with total system fluid volume. Here, a comprehensive review of single-phase heat transfer enhancement techniques is presented. The article provides a thorough comparison by analyzing the heat transfer rate, pressure drop, and other operational aspects. Single-phase heat transfer enhancement methods are divided into active and passive techniques. Active methods such as electrohydrodynamic (EHD), magnetohydrodynamics (MHD), or mechanical motion require external power to create enhancement. Passive methods such as dimples, fins, or tape inserts do not require external input and rely only on surface modification. Although active methods are more expensive and difficult to implement compared to passive techniques, it enables active control of heat transfer augmentation. Finally, this review develops and summarizes key learning data for design optimization enabled by additive manufacturing and machine learning algorithms, helping to inform these next-generation heat exchanger design methodologies for a plethora of modern applications such as electrification of vehicles, computing, and classical industries.

42 ENGINEERING↗
DOE OSTI2023

An Iterative Bidirectional Gradient Boosting Algorithm for CVR Baseline Estimation

This paper presents a novel iterative, bidirectional, gradient boosting (bidirectional-GB) algorithm for estimating the baseline of the Conservation Voltage Reduction (CVR) program. We define the CVR baseline as the load profile during the CVR period if the substation voltage is not lowered. The proposed algorithm consists of two key steps: selection of similar days and iterative bidirectional-GB training. In the first step, preand post-event temperature profiles of the targeted CVR day are used to select similar days from historical non-CVR days. In the second step, the pre-event and post-event similar days are used to train two GBMs iteratively: a forward-GBM and a backwardGBM. After each iteration, the two generated CVR baselines are reconciled and only the first and the last points on the reconciled baseline are kept. The iteration repeats until all CVR baseline points are generated. We tested two gradient boosting methods (i.e., GBM and LighGBM) with two data resolutions (i.e., 15and 30-minute). The results demonstrate that both the accuracy and performance of the algorithm are satisfactory.

Lee, Han Pyo↗
DOE OSTI2024

Machine Learning for Joint Quality Control

The use of lightweight material combinations has been highly demanded in manufacturing automotive structures. However, making robust dissimilar material joints of such lightweight materials is still challenging. A significant barrier to achieving high-quality and repeatable joint performance is a deficient understanding of the relationship between the welding process, joint attributes, and joint performance. In this context, welding factors refer to material, equipment, environment, and process parameters, while joint features comprise specific microstructural attributes of the weld such as nugget size, heat affected zone (HAZ) topology, intermetallic layer thickness, and sheet thickness reduction. Joint performance is quantified in terms of strength (e.g., tensile shear, coach peel, cross-tension), weld size, and hardness, among other factors. While there have been many attempts to establish this process-structure-property relationship by developing a model derived from the associated physics and first principles, the complexity of the joining processes compounded by the complex interactions with different materials in an automotive assembly line environment, has hindered the usefulness of such attempts. The complexity is further exacerbated using different stacking materials, especially comprising dissimilar material combinations. In practice, the common approach has been the laborious process of creating welds, characterizing them, and then physically testing them through experimentation. With the emergence of artificial intelligence (AI) methods, an alternative pathway to eliciting the desired process-structure-property relationship at an accelerated pace is to use a data-driven approach by employing machine-learning (ML) techniques. This approach is benefitted by the availability of large streams of data, generated through years of research and testing by original equipment manufacturers, in the form of material, process, environmental, equipment, microstructural, and bulk-scale performance information from multimodal, multiscale sensors making measurements from laboratory-scale to production-scale processes. During Phase I efforts, which ended in fiscal year (FY) 2021, the Oak Ridge National Laboratory and Pacific Northwest National Laboratory (ORNL/PNNL) team demonstrated the effectiveness of different ML/AI frameworks in modeling complex relationships between resistance spot welding (RSW) process parameters, weld attributes, and joint properties using a subset of data from General Motors (GM). In FY 2022, the project team further refined and expanded their respective ML models to analyze additional welds with new weld stack-ups and materials to enhance the ML model predictive capability. ORNL extended its unified deep neural networks (DNN) ML training and prediction framework with new data streams of process parameters, and PNNL extended its model describing RSW process parameters’ associations with weld attributes. In FY 2023, the project team completed the development of the AI/ML architecture for analyzing aluminum/steel joints manufactured by GM via RSW and transitioned into the inline welding quality monitoring task for steel/steel RSW joints provided by GM.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Accelerating Flash-X Simulations with Asynchronous I/O

Most high-fidelity physics simulation codes, such as Flash-X, need to save intermediate results (checkpoint files) to restart or gain insights into the evolution of the simulation. These simulation codes save such intermediate files synchronously, where computation is stalled while the data is written to storage. Depending on the problem size and computational requirements, this file write time can be a substantial portion of the total simulation time. In order to hide the I/O latency of checkpointing, asynchronous I/O methods have been introduced. These methods use background threads for performing I/O while the main threads continue with the simulation. The usage of background threads can compete for resources on the node as well as with communication. In this paper, we evaluate the overheads and the overall benefit of asynchronous I/O in HDF5 to simulations. Results from real-world high-fidelity simulations on the Summit supercomputer show that I/O operation is overlapped with application communication or computation or both, effectively hiding some or all of the I/O latency. Our evaluation shows that while using asynchronous I/O adds overhead to the application, the I/O time reduction is more significant, resulting in overall up to 1.5X performance speedup.

Jain, Rajeev↗
DOE OSTI2021

Transient knockdown of Anopheles stephensi LRIM1 using RNAi increases Plasmodium falciparum sporozoite salivary gland infections

Abstract Background Plasmodium falciparum (Pf) sporozoites (PfSPZ) can be administered as a highly protective vaccine conferring the highest protection seen to date. Sanaria® PfSPZ vaccines are produced using aseptically reared Anopheles stephensi mosquitoes. The bionomics of sporogonic development of P. falciparum in A. stephensi to fully mature salivary gland PfSPZ is thought to be modulated by several components of the mosquito innate immune system. In order to increase salivary gland PfSPZ infections in A. stephensi and thereby increase vaccine production efficiency, a gene knock down approach was used to investigate the activity of the immune deficiency (IMD) signaling pathway downstream effector leucine-rich repeat immune molecule 1 ( LRIM1 ), an antagonist to Plasmodium development. Methods Expression of LRIM1 in A. stephensi was reduced following injection of double stranded (ds) RNA into mosquitoes. By combining the Gal4/UAS bipartite system with in vivo expression of short hairpin (sh) RNA coding for LRIM1 reduced expression of LRIM1 was targeted in the midgut, fat body, and salivary glands. RT-qPCR was used to demonstrate fold-changes in gene expression in three transgenic crosses and the effects on P. falciparum infections determined in mosquitoes showing the greatest reduction in LRIM1 expression. Results LRIM1 expression could be reduced, but not completely silenced, by expression of LRIM1 dsRNA. Infections of P. falciparum oocysts and PfSPZ were consistently and significantly higher in transgenic mosquitoes than wild type controls, with increases in PfSPZ ranging from 2.5- to tenfold. Conclusions Plasmodium falciparum infections in A. stephensi can be increased following reduced expression of LRIM1. These data provide the springboard for more precise knockout of LRIM1 for the eventual incorporation of immune-compromised A. stephensi into manufacturing of Sanaria’s PfSPZ products.

Billingsley, Peter F. (ORCID:0000000311424813)↗
DOE OSTI2020

The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules

Abstract Maximum diversification of data is a central theme in building generalized and accurate machine learning (ML) models. In chemistry, ML has been used to develop models for predicting molecular properties, for example quantum mechanics (QM) calculated potential energy surfaces and atomic charge models. The ANI-1x and ANI-1ccx ML-based general-purpose potentials for organic molecules were developed through active learning; an automated data diversification process. Here, we describe the ANI-1x and ANI-1ccx data sets. To demonstrate data diversity, we visualize it with a dimensionality reduction scheme, and contrast against existing data sets. The ANI-1x data set contains multiple QM properties from 5 M density functional theory calculations, while the ANI-1ccx data set contains 500 k data points obtained with an accurate CCSD(T)/CBS extrapolation. Approximately 14 million CPU core-hours were expended to generate this data. Multiple QM calculated properties for the chemical elements C, H, N, and O are provided: energies, atomic forces, multipole moments, atomic charges, etc. We provide this data to the community to aid research and development of ML models for chemistry.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Adapting In Situ Accelerators for Sparsity With Granular Matrix Reordering

Neural network (NN) inference is an essential part of modern systems and is found at the heart of numerous applications ranging from image recognition to natural language processing. In situ NN accelerators can efficiently perform NN inference using resistive crossbars, which makes them a promising solution to the data movement challenges faced by conventional architectures. Although such accelerators demonstrate significant potential for dense NNs, they often do not benefit from sparse NNs, which contain relatively few non-zero weights. Processing sparse NNs on in situ accelerators results in wasted energy to charge the entire crossbar where most elements are zeros. To address this limitation, this paper proposes Granular Matrix Reordering (GMR): a preprocessing technique that enables an energy-efficient computation of sparse NNs on in situ accelerators. GMR reorders the rows and columns of sparse weight matrices to maximize the crossbars' utilization and minimize the total number of crossbars needed to be charged. The reordering process does not rely on sparsity patterns and incurs no accuracy loss. Finally, GMR achieves an average of 28% and up to 34% reduction in energy consumption over seven pruned NNs across four different pruning methods and network architectures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Invertible Neural Networks for Aerodynamic Design of Wind Turbine Blades

The state-of-the-practice methods for aerodynamic design of wind turbine blads use linearized blade element momentum theory (BEM) to optimize the twist and chord profiles from a pre-selected set of 2D airfoil shapes. In this work, we apply invertible neural network (INN) tools to enable the rapid inverse aerodynamic design of wind turbine blades including component airfoils. The INN is trained on data obtained through the use of robust automated mesh generation and the HAMSTR computational fluid dynamics solver with advanced turbulence and transition models validated for turbine applications. Our design technique is a significant improvement over the state-of-the-practice linearized blade element momentum (BEM) techniques in capturing 3D nonlinear aerodynamic effects that are critical for optimal design of the rotors. This is made possible by developing sparse, invertible neural networks (INNs) for inverse design and optimization that realize a 100x cost reduction compared to adjoint-based computational fluid dynamics (CFD) approaches, while enabling increased robustness of the final design. We demonstrate the INN tool for design of a section of the NREL 5-MW blade. All generated shapes satisfy the desired aerodynamic characteristics, demonstrating the success of the INN approach for inverse design of wind turbine blades.

aerodynamic design↗