Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Generative Design”

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 1,693 records · Page 94
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Spacetime pq theory for AC and DC electric power systems

The 50/60 Hz alternating current (AC) electric power has been the standard and most flexible energy source powering our modern societies for one and a half centuries since the war of the currents: AC versus direct current (DC). A reactive power concept that was introduced at the beginning of the AC power was very useful for circuit/system analysis, design, control, optimization, and ultimately for more efficient and stable generation, transmission, distribution, and consumption. The initial reactive power theory was based on single-phase sinusoidal AC power to capture inductive and capacitive power that yields to net-zero average power over one fundamental cycle. Soon it was expanded to non-sinusoidal AC power and finally to instantaneous three-phase AC power. However, these reactive power theories remain separate and limited to special cases and have never been consolidated and made valid to all cases. Today, more widespread adoption of power electronics and renewable energy is bringing back DC power into the electric grids. The reactive power concept has never been applied to DC power systems. There is no reactive power in DC power systems according to the existing reactive power theories. Do DC power systems really have no reactive power? Capacitors and inductors are widely used in DC just like in AC power systems. Are they not reactive power components? Why are they different from their AC counterparts? Furthermore, are batteries active or reactive power components? What about active devices like power converters (or inverters) with AC (or DC) on one side and DC (or AC) on the other? Do they generate or consume reactive power? Finally, what about AC and DC hybrid power systems? How to define reactive power in such a complex power system that has a multitude of loads, buses, and sources? Is there reactive power between any two loads, any two buses, or any two sources in a power system and what is the total reactive power in such a complex power system as a whole? As the motivation and goal of this paper to answer the above basic questions, to unify the existing AC reactive power theories and to ultimately provide theoretical and insightful guidance for system analysis, design, control, efficiency, optimization, and operation of complex power systems, a concept of spacetime (both spatial and temporal) active and reactive power (pq) theory—the spatiotemporal aspect of active and reactive power—is developed for both AC and DC power systems. The theoretical definitions and physical meanings of the spacetime reactive power will be developed, and real applications and thought experiments/cases/exercises will be explored and discussed. The developed mathematics to define the active (or real) and reactive (or imaginary) power— p and q respectively by dot (scalar) and cross (vector) products of multi-dimension spacetime vectors and time-space mapping principle/law can have some fundamental implications as well.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2021

APS: High-Energy X-rays Expediting Applied and Fundamental Research

The Advanced Photon Source (APS) at Argonne National Laboratory (ANL) was designed and built as one of the first large 3rd generation synchrotrons operated at 7 GeV. The United States Department of Energy (DOE) Office of Science funded the construction and continues to fund the operation of this national facility. Furthermore, the beamlines were originally operated by Collaborative Access Teams (CAT), which allowed a strong involvement of research institutes, universities, and industry. Today 43 of 68 beamlines are operated by the X-ray Science Division (XSD) and 35 beamlines are directly funded by the DOE Office of Science. The APS became operational in 1995, and was planned right from the beginning with 4 beamlines dedicated to high-energy x-rays above 40 keV: 1-ID designed to study engineering materials science, 6-ID-D for magnetism and condensed matter research, 11-ID-B dedicated to magnetic Compton scattering using an Elliptical Multipole Wiggler, and 11-ID-C for materials science and condensed matter physics. These stations were later joined by the bending magnet beamlines 6-BM for white beam studies and 17-BM for rapid powder diffraction. CAT operated beamlines such as 15-ID for crystallography or 16-ID for high-pressure studies also use the high-energy x-ray regime frequently. The program in Compton scattering ended in 2004 and the wiggler was replaced by an undulator. 11-ID-B became the first dedicated station for Pair Distribution Function (PDF) Analysis.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Utilization of ACE nuclear data file toolkit ACEtk to calculate relative sensitivity coefficients of point-kinetics parameters

Sensitivity and uncertainty methods are quintessential for nuclear criticality safety and experiment design. This type of analysis relies on calculations of sensitivity coefficients; sensitivity coefficients of the effective neutron multiplication factor with respect to some nuclear data are predominantly calculated and used. As a part of the Laboratory Directed Research & Development project EUCLID (Experiments Underpinned by Computational Learning for Improvements in nuclear Data) at Los Alamos National Laboratory, sensitivity coefficients of many radiation detector measurement responses with respect to nuclear data were investigated. Specifically, this paper outlines a method to calculate point-kinetics parameters relative sensitivity coefficients with respect to nuclear data. Point-kinetics parameters such as the prompt neutron decay constant, effective delayed neutron fraction, and neutron generation time are especially important to experimenters and reactor operators designing systems with dynamic neutron populations. This method couples capabilities of the ACE (A Compact ENDF) nuclear data file toolkit, ACEtk, with the ability to load cross sections into the radiation transport code Monte Carlo N-Particle (MCNP). In conclusion, key aspects of optimizing this method for a particular application and sensitivity profiles of the Jezebel criticality experiment are examined and discussed.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2024

Quapprox: A Framework for Benchmarking the Approximability of Variational Quantum Circuit

Most of the existing quantum neural network models, such as variational quantum circuits (VQCs), are limited in their ability to explore the non-linear relationships in input data. This gradually becomes the main obstacle for it to tackle realistic applications, such as natural language processing, medical image processing, and wireless communications. Recently, there have emerged research efforts that enable VQCs to perform non-linear operations. However, it is still unclear on the approximability of a given VQC (i.e., the order of non-linearity that can be handled by a specified design). In response to this issue, we developed an automated tool designed to benchmark the approximation of a given VQC. The proposed tool will generate a set of synthetic datasets with different orders of non-linearity and train the given VQC on these datasets to estimate their approximability. Our experiments benchmark VQCs with different designs, where we know their theoretic approximability. We then show that the proposed tool can precisely estimate the approximability, which is consistent with the theoretic value, indicating that the proposed tool can be used for benchmarking the approximability of a given quantum circuit for learning tasks.

artificial intelligence↗
DOE OSTI2021

An Economics-by-Design Approach Applied to a Heat Pipe Microreactor Concept

Microreactors present a potential paradigm shift in the nuclear industry. Emphasis thus far has been on large-scale multi-billion-dollar projects that cater solely to grid electricity market. These projects can be challenging to finance and execute. On the other hand, microreactors are intended to target a wide variety of smaller niche markets and are expected to be factory-fabricated and more readily deployable. While diseconomies of scale for microreactors may tend to raise their costs per energy output (MWh) relative to large nuclear plants, offsetting gains can be expected from standardization, simplification, passive safety, lower radionuclide inventories, factory fabrication, fast installation, and low financing costs. To adequately assess these contributions, designers should have a different perspective on cost drivers than for large nuclear plants and can utilize novel approaches for systematic cost reduction. To account for these important aspects of microreactors, this report proposes an economics-by-design approach that places economic considerations at the center of the design process. The methodology builds on existing frameworks such as design-to-cost and value engineering, expanding them to new markets (beyond the grid), new attributes (beyond costs alone), and introducing the approach at earlier points in the design cycle. Design parameters and technical specifications are systematically evaluated until costs meet market entry points, while also providing the high-priority performance attributes of the particular use case. Determining first-order estimates for different components early in the process enables designers to focus R&D efforts on the biggest overall cost contributors and components with the most cost uncertainty. The analysis is always guided by market needs and threshold prices. In addition to microreactors, the approach is expected to be useful for other classes of nuclear reactors as well. The analysis was applied to a concept found in the open literature (the Design A heat-pipe reactor). A comprehensive bottom-up estimate was generated by leveraging a new microreactor-specific code of accounts and a range of cost equations. The initial estimate for levelized cost of electricity (LCOE) unsurprisingly exceeded market ranges since the use case had prioritized technological readiness over economic considerations in design choices. An alternate concept was then proposed, with various assumptions/targets made to reduce the largest cost contributors. Changes in the neutron spectrum, the power output, and building structures were found to make even the first-of-a-kind of this modified concept competitive with diesel generation in some remote communities. Learning rate (LR) assumptions indicated cost reductions achieved from sequential unit deployments could expand the range of competitiveness to include additional markets as deployments proceed.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Design of biphenylene-derived tunable dirac materials

The exploration of carbon allotropes has unveiled a series of two-dimensional (2D) materials with unique electronic and mechanical properties, yet the need for stable structures with tailored electronic properties persists. Here, in this study, we introduce a new class of 2D carbon allotropes derived from the biphenylene network (BPN), incorporating acetylenic linkages to tune their structural and electronic characteristics. Through density functional theory calculations, we identified ten novel BPN-derived structures that exhibit both energetic and dynamic stability, confirmed by cohesive energy and phonon spectrum analyses. Among them, BPN-02 and BPN04 are metallic, featuring critically-tilted type-III Dirac cones under ~ 5 % biaxial strain, while BPN-22 is a semiconductor with a band gap of 0.95 eV and exhibits highly anisotropic carrier mobility. Additionally, these structures demonstrate significant anisotropy in their elastic properties, further distinguishing them from other 2D carbon materials like graphene. Our findings suggest that these novel BPN-based structures have strong potential for next-generation electronic and optoelectronic applications, providing new avenues for the design and synthesis of advanced carbon materials.

2D carbon allotropes↗
DOE OSTI2025

Design of an Out-Of-Pile Experimental Facility to Demonstrate the Feasibility of In Situ Thermal Conductivity Measurements of Nuclear Fuels Under Irradiation

There is substantial merit in quantifying nuclear fuel performance under irradiation. At Oak Ridge National Laboratory (ORNL), the MiniFuel irradiation platform has become the primary test vehicle for conducting separate-effects fuel performance irradiation experiments. The MiniFuel experiment is a passively controlled capsule design deployed in the High Flux Isotope Reactor (HFIR) through which fuel performance data is collected post-irradiation. Separate effects fuels irradiation capabilities are being expanded at ORNL by developing instrumented capsule designs that aim to capture fuel performance phenomena in-situ. One such capsule will specifically target fuel specimen thermal conductivity changes as a function of fuel burnup. Due to the complexity of making this measurement on nuclear fuel in-pile, this paper describes the necessary out-of-pile testing conducted on the thermal conductivity capsule (TCC) design. The measurement is ascertained via a thermopile system with heat transferred unidirectionally through a surrogate fuel specimen sandwiched between two conductive materials. The capsules investigated in this study are representative of the in-pile design, with the primary departure from irradiation conditions being the distribution of heat generation within the capsule. In the out-of-pile experiment, an external heater was used to drive heat through the conductive slug materials and into the specimen. This paper expounds the design of the out-of-pile experimental system and the thermal conductivity measurement technique. Predictive models used to determine the sensitivity of the measurement to variables governing thermal contact conductance between the specimen and slug materials and to predict experimental results are also described. Data from the out-of-pile experiment will be used to validate the readiness of the design for insertion into HFIR for irradiation.

Parker, Trevor [ORNL]↗
DOE OSTI2022

Illinois Compressed Air Energy Storage

Compressed Air Storage Energy (CAES) is one of the few mid- technology readiness level (TRL) energy storage technologies that can address the long-duration infrastructure needed for dealing with variable electric output from renewable energy sources and be reliable backup source for replacing natural gas during supply interruptions. In CAES the goal is to capture and store compressed air in subsurface sedimentary strata when off-peak power is available, or there is a need for grid balancing. The stored high-pressure air is returned to the surface and used to power turbines during reductions in either renewable energy or supply issues with fossil fuels. The Illinois CAES project evaluates the feasibility of capturing surplus electrical energy from renewable sources and off-peak energy at a fossil fuel power plant at the University of Illinois Urbana - Champaign (UIUC) campus. The UIUC Abbott Power Plant uses natural gas and coal to generate electricity (capacity: 35 MWe by coal and 49 MWe by NG). UIUC receives additional electricity from on campus solar farm, and off-campus wind farm. Also, UIUC offsets electricity usage by integrating geothermal energy systems into building heating Also, UIUC offsets steam, hot and chilled water usage by integrating geothermal energy systems into building heating and cooling systems. Furthermore, the two UIUC solar farms (Solar Farm 1 is 21 acres and Solar Farm 2 is 54 acres) to generate 4.68 megawatts (MW) and 12.1 MW, respectively. Campus receives 8.6% of the wind-generated electricity from the Rail Splitter Wind Farm. The project objectives were to design an integrated system to 1) capture surplus electrical energy from renewable sources and the Abbott Power Plant using a CAES system, 2) store both the compressed air and the thermal heat generated by compression in the subsurface as part of an adiabatic system, 3) simulate the movement of the air and heat in the subsurface, 4) recover the compressed air and stored thermal heat to rotate turbine generators during sustained interruption due to weather events or fossil fuel disruptions.

03 NATURAL GAS↗
DOE OSTI2024

Development of an Optimal Variable-Pitch Controller for Floating Axial-Flow Marine Hydrokinetic Turbines

This article discusses the development of an optimal variable-pitch controller for floating, axial-flow marine turbines. Recently, OpenFAST, an open-source wind turbine modeling tool, has been extended to model marine turbines. A controller is necessary to simulate marine turbines for different load cases using OpenFAST, which greatly impacts the performance of the energy system. Previous studies have designed controllers using a linearized model of the marine turbine, which can be time-consuming and require the expertise of a control engineer. In this study, we use an automated approach that uses generic models of the marine turbine to identify the controller gains, which can expedite the process of designing a controller. Using an optimizer to identify the control system parameters can additionally improve the controller's performance. The optimal controller tuned using such an approach results in a 20% reduction in the tower-base damage equivalent loading and better tracking of the rated generator speed and power.

axial flow↗
DOE OSTI2024

Twentieth Exotic Beam Summer School (EBSS2023)

The study of unstable nuclei with unusual ratios of protons to neutrons is one of the frontiers of science. Investigating these rare isotopes is critical for understanding the synthesis of the chemical elements in stellar explosions as well as the fundamental nature of the nuclear forces that bind atomic nuclei together. Scientific progress in this field is driven by the development of exotic beams in present and next-generation rare-isotope beam facilities including the Facility for Rare Isotope Beams (FRIB). Nuclear physics is a broad discipline, influencing our knowledge on subjects as diverse as weakly-bound nuclei, many-body quantum theory, the super heavy elements, and the inner structure of neutron stars. Applications based on nuclear science and technologies include medical diagnostics and therapies, materials science, and national security. The major goal achieved in this project was to hold Exotic Beam Summer School 2023 (EBSS2023), the twentieth installment of EBSS series, July 9-15, 2023 at the Facility for Rare Isotope Beams on the campus of Michigan State University to educate and train the next generation of scientists that will drive research with rare-isotope beams. FRIB became operational in 2022 and is now providing beams of exotic nuclei that will ramp up to unmatched intensities, exceeding what is available today by orders of magnitude. Beams available at FRIB are facilitating a wide variety of studies in nuclear structure, astrophysics, fundamental symmetries and societal applications. There is a large community of scientists interested in working with rare isotope beams; for example, the FRIB User Organization currently has over 1,700 members. In order to maximize the scientific output of FRIB, there must be a workforce continuously trained in both the physics of exotic beams and in the practical techniques of carrying out an experiment. This summer school series is designed to specifically address this need - to ensure that new generations of scientists from a broad range of institutions and backgrounds is trained, motivated, and equipped to push the field forward to new and important breakthroughs.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

A Survey of Machine Learning-Based Physics Event Generation

Event generators in high-energy nuclear and particle physics play an important role in facilitating studies of particle reactions. We survey the state of the art of machine learning (ML) efforts at building physics event generators. We review ML generative models used in ML-based event generators and their specific challenges, and discuss various approaches of incorporating physics into the ML model designs to overcome these challenges. Finally, we explore some open questions related to super-resolution, fidelity, and extrapolation for physics event generation based on ML technology.

Alanazi, Yasir↗
DOE OSTI2022

Lighting application efficacy: A framework for holistically measuring lighting use in buildings

Lighting consumes significant energy in buildings, but current measurements of lighting efficiency are inadequate in capturing the spatial and temporal effectiveness of light. The most widely used lighting energy efficiency measure, luminous efficacy, quantifies the amount of light generated by individual light sources. While luminous efficacy provides manufacturers and designers with a useful metric for quantifying the energy efficiency of individual lighting devices, it does not fully address the use of lighting in architectural spaces. Instead, application efficacy, the relationship between the electrical power consumed by lighting hardware and the amount of light that contributes to the visual perception of building occupants, must be measured. Here, we propose a framework for calculating lighting application efficacy (LAE) based on the primary pathway of light in architectural settings: the generation and emission of light from a luminaire, the travel of the light throughout the space and into occupants’ eyes, and the process of visual perception. The LAE framework will include the development of computational methods for estimating the proportion of emitted light that is directed to task areas and areas corresponding to occupants’ visual fields using ray-tracing lighting design software. Future modeling and validation work will quantify energy efficiency of buildings.

Durmus, Dorukalp↗
DOE OSTI2024

ABC -Auxetics: An Implicit Design Approach for Negative Poisson's Ratio Materials

A novel methodology is introduced for designing auxetic (negative Poisson's ratio) structures based on topological principles and is demonstrated by investigating a new class of auxetics based on two-dimensional (2D) textile weave patterns. Conventional methodology for designing auxetic materials typically involves determining a single deformable block (a unit cell) of material whose shape results in auxetic behavior. Consequently, patterning such a unit cell in a 2D (or 3D) domain results in a larger structure that exhibits overall auxetic behavior. Such an approach naturally relies on some prior intuition and experience regarding which unit cells may be auxetic. Second, tuning the properties of the resulting structures is typically limited to parametric variations of the geometry of a specific type of unit cell. Thus, most of the currently known auxetic structures belong to a selected few classes of unit cell geometries that are explicitly defined in accordance with a specified topological (i.e., grid structure). Herein, a new class of auxetic structures is demonstrated that, while periodic, can be generated implicitly, i.e., without reference to a specific unit cell design. The approach leverages weave-based parameters (A–B–C), resulting in a rich design space for auxetics that is previously unexplored.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Pickaxe: a Python library for the prediction of novel metabolic reactions

Abstract Background Biochemical reaction prediction tools leverage enzymatic promiscuity rules to generate reaction networks containing novel compounds and reactions. The resulting reaction networks can be used for multiple applications such as designing novel biosynthetic pathways and annotating untargeted metabolomics data. It is vital for these tools to provide a robust, user-friendly method to generate networks for a given application. However, existing tools lack the flexibility to easily generate networks that are tailor-fit for a user’s application due to lack of exhaustive reaction rules, restriction to pre-computed networks, and difficulty in using the software due to lack of documentation. Results Here we present Pickaxe, an open-source, flexible software that provides a user-friendly method to generate novel reaction networks. This software iteratively applies reaction rules to a set of metabolites to generate novel reactions. Users can select rules from the prepackaged JN1224min ruleset, derived from MetaCyc, or define their own custom rules. Additionally, filters are provided which allow for the pruning of a network on-the-fly based on compound and reaction properties. The filters include chemical similarity to target molecules, metabolomics, thermodynamics, and reaction feasibility filters. Example applications are given to highlight the capabilities of Pickaxe: the expansion of common biological databases with novel reactions, the generation of industrially useful chemicals from a yeast metabolome database, and the annotation of untargeted metabolomics peaks from an E. coli dataset. Conclusion Pickaxe predicts novel metabolic reactions and compounds, which can be used for a variety of applications. This software is open-source and available as part of the MINE Database python package ( https://pypi.org/project/minedatabase/ ) or on GitHub ( https://github.com/tyo-nu/MINE-Database ). Documentation and examples can be found on Read the Docs ( https://mine-database.readthedocs.io/en/latest/ ). Through its documentation, pre-packaged features, and customizable nature, Pickaxe allows users to generate novel reaction networks tailored to their application.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Innovating the next generation of commercial smart building software

Nearly 30% of commercial building energy use is wasted due to equipment faults and HVAC controls problems. The result is increased emissions, compromised comfort and productivity, and less reliable coordination of building power needs with a clean grid. The energy impact alone represents $17 billion in potential savings. Today’s smart building software provides a robust solution to address these operational deficiencies. Energy management and information systems (EMIS) are saving up to 9% on average, with two-year paybacks. They are being incorporated into energy management processes, commissioning services, and utility programs. As effective as they are, two barriers prevent even deeper benefits; limited personnel to fix problems once they are identified, and the expense and time to manually implement changes in control systems. In partnership with the research community, the EMIS industry is developing new capabilities to overcome these barriers. Moving beyond siloed products for either fault detection and diagnostics, or optimal control, these new capabilities empower users to not only automatically identify faults, but also to push corrective action, and control improvements to their buildings. In this paper, several areas for enhancements are documented: ‘one-time’ correction of faults such as setpoints, schedules, and economizer lockouts; short-term active testing for automated proportional integral derivative (PID) loop tuning and functional testing; and continuous supervisory control for demand flexibility and year-round efficiency. Results are presented from a pair of partner implementations out of a dozen providers integrating these enhancements into their products, including field tests from across the country, and insights into operator acceptance and integration into operations and maintenance practices.

Casillas, Armando↗
DOE OSTI2020

Incorporating the Impacts of Climate Change on Hydrology in a Performance Assessment Model - 20403

The evidence of climate change is increasingly well-documented and impacts should be incorporated in performance assessment studies. The current climate literature provides both observational evidence and climate model projections of climate trends and/or climate change in the late 20. and early 21. centuries for North America and the northeast United States. Probabilistic modeling is a core requirement for quantifying uncertainty and evaluating its impacts. Not evaluating future climate states in a performance assessment because of the existence of uncertainty is contradictory to good modeling practices - the most uncertain issues and parameters require the most attention in effective probabilistic modeling. Excluding climate change limits development of modeling information that could aid in effective decision making. In this work we develop methods to use the output from hydrologic models and analysis of historical aerial imagery to quantify and implement the impacts of climate change on hydrologic processes at a nuclear waste site in West Valley, New York. Specifically, we used the HELP (Hydraulic Performance of Landfill Performance) model to characterize key hydrologic processes under both current and future climate conditions to assess the impacts of changing climate on hydrology. A suite of previous climatic models were reviewed and synthesized to produce a cohesive representation of the current state of knowledge of the impact of climate change on important model inputs such as precipitation. Output from the HELP simulations was coupled to the GoldSim model that was used to develop the Probabilistic Performance Assessment (PPA) approach through the application of a novel 'nearest neighbor' technique. First, several thousand realizations were generated from the HELP model using a Latin Hypercube experimental design to ensure adequate coverage of the parameter space of explanatory variables used to drive HELP. For example, porosity is a physical parameter that is used as an input to both HELP and the GoldSim PA model. We then ran sensitivity analysis (SA) algorithms on the output of HELP for each of the responses of interest. For each predictor, each time we build an SA model we get a different value for the sensitivity index (SI). From the collection of all the SA models, the average was computed among all of the SIs to represent the predictor within the context of the nearest neighbor approach. That is, we conduct SA on each HELP outcome for each scenario. This gives us parameter sensitivity indices for the outcomes. We average the parameter sensitivity indices across the outcomes to get the average SI for a scenario. For each realization that is generated from the Goldsim PA model, Goldsim generates random values for physical/empirical parameters that HELP uses as well. For each vector of physical/empirical parameters that Goldsim generates, the vector from the 5,000 HELP runs that is most 'similar' to the Goldsim vector is computed using the nearest neighbor approach. In this context 'similar' means minimization of the SA-weighted sum of the absolute differences among the 5,000 values computed for this statistic, where each value corresponds to a different HELP realization. In order to account for the impacts of climate change, this process was repeated using the spatially downscaled future climate projections. For each of the key parameters of interest, it was assumed that a linear change depicted the relationship between the values for the present day and those for 2100. In this way, the climatically-driven changes in key parameters used to inform the GoldSim model are quantified and incorporated into the PA model output for the future. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2022

Invertible Neural Networks for Airfoil Design

We report the airfoil design problem, in which an engineer seeks a shape with desired performance characteristics, is fundamental to aerodynamics. Design workflows traditionally rely on iterative optimization methods using low-fidelity integral boundary-layer methods as higher-fidelity adjoint-based computational fluid dynamics methods are computationally expensive. Surrogate-based approaches can accelerate the design process but still rely on some iterative inverse design procedure. In this work, we leverage emerging invertible neural network (INN) tools to enable the rapid inverse design of airfoil shapes for wind turbines. INNs are specialized deep-learning models with well-defined inverse mappings. When trained appropriately, INN surrogate models are capable of forward prediction of aerodynamic and structural quantities for a given airfoil shape as well as inverse recovery of airfoil shapes with specified aerodynamic and structural characteristics. The INN approach offers a roughly 100 times speed-up compared to adjoint-based methods for inverse design. We demonstrate the INN tool for inverse design on three test cases of 100 airfoils each that satisfy the performance characteristics close to those of airfoils used in wind-turbine blades. All generated shapes satisfy the desired aerodynamic characteristics, demonstrating the success of the INN approach for inverse design of airfoils.

17 WIND ENERGY↗