SEARCH · Search NASA
Results for “Reduced Order Modeling”
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.
A fundamental theorem for the model reduction of nonlinear systems
A simple, but fundamental, theorem is given on the extent to which a nonlinear system model can have its order reduced. Essentially, the result is that the order, or the dimension of the state space representation, cannot be reduced to, or below, the dimension of the system's attractor. Several examples are given to illustrate this point. The result is especially applicable to higher order systems such as the infinite dimensional systems arising from the modeling of distributed parameter systems.
Fast Approximate Broadband Phase Retrieval for Segmented Systems
Broadband phase retrieval needed when: a) Narrow spectral filters are unavailable. b) Dim sources. c) Low throughput due to misalignment. d) Short exposures times. i.e., Pointing instability (space); and Atmospheric instability (ground based AO). Traditional approach is computationally burdensome for extreme bandwidths. Approximate approach: a) Substitute monochromatic model. b) Blur model and data. Test case performance: a) approx.270x reduction in computational cost for FGS-like test case. b) Good accuracy for monolithic system. c) Acceptable accuracy for segmented systems. i.e., Reduced by diffraction and Reduced by higher order segment model.
BUQEYE guide to projection-based emulators in nuclear physics
The BUQEYE collaboration (Bayesian Uncertainty Quantification: Errors in Your effective field theory) presents a pedagogical introduction to projection-based, reduced-order emulators for applications in low-energy nuclear physics. The term emulator refers here to a fast surrogate model capable of reliably approximating high-fidelity models. As the general tools employed by these emulators are not yet well-known in the nuclear physics community, we discuss variational and Galerkin projection methods, emphasize the benefits of offline-online decompositions, and explore how these concepts lead to emulators for bound and scattering systems that enable fast and accurate calculations using many different model parameter sets. We also point to future extensions and applications of these emulators for nuclear physics, guided by the mature field of model (order) reduction. All examples discussed here and more are available as interactive, open-source Python code so that practitioners can readily adapt projection-based emulators for their own work.
Transforming Energy Through Computational Excellence: Co-Optimized Machine-Learned Manifold Models for Large Eddy Simulation of Turbulent Combustion
Summary of the challenge, approach, results, and impact of co-optimized machine-learned manifold models for research of large eddy simulation of turbulent combustion.
Core Noise - Increasing Importance
This presentation is a technical summary of and outlook for NASA-internal and NASA-sponsored external research on core (combustor and turbine) noise funded by the Fundamental Aeronautics Program Subsonic Fixed Wing (SFW) Project. Sections of the presentation cover: the SFW system-level noise metrics for the 2015, 2020, and 2025 timeframes; turbofan design trends and their aeroacoustic implications; the emerging importance of core noise and its relevance to the SFW Reduced-Perceived-Noise Technical Challenge; and the current research activities in the core-noise area, with additional details given about the development of a high-fidelity combustor-noise prediction capability as well as activities supporting the development of improved reduced-order, physics-based models for combustor-noise prediction. The need for benchmark data for validation of high-fidelity and modeling work and the value of a potential future diagnostic facility for testing of core-noise-reduction concepts are indicated. The NASA Fundamental Aeronautics Program has the principal objective of overcoming today's national challenges in air transportation. The SFW Reduced-Perceived-Noise Technical Challenge aims to develop concepts and technologies to dramatically reduce the perceived aircraft noise outside of airport boundaries. This reduction of aircraft noise is critical to enabling the anticipated large increase in future air traffic. Noise generated in the jet engine core, by sources such as the compressor, combustor, and turbine, can be a significant contribution to the overall noise signature at low-power conditions, typical of approach flight. At high engine power during takeoff, jet and fan noise have traditionally dominated over core noise. However, current design trends and expected technological advances in engine-cycle design as well as noise-reduction methods are likely to reduce non-core noise even at engine-power points higher than approach. In addition, future low-emission combustor designs could increase the combustion-noise component. The trend towards high-power-density cores also means that the noise generated in the low-pressure turbine will likely increase. Consequently, the combined result from these emerging changes will be to elevate the overall importance of turbomachinery core noise, which will need to be addressed in order to meet future noise goals.
Krylov vector methods for model reduction and control of flexible structures
Krylov vectors and the concept of parameter matching are combined here to develop model-reduction algorithms for structural dynamics systems. The method is derived for a structural dynamics system described by a second-order matrix differential equation. The reduced models are shown to have a promising application in the control of flexible structures. It can eliminate control and observation spillovers while requiring only the dynamic spillover terms to be considered. A model-order reduction example and a flexible structure control example are provided to show the efficacy of the method.
Modeling household online shopping demand in the U.S.: a machine learning approach and comparative investigation between 2009 and 2017
Despite the rapid growth of online shopping and research interest in the relationship between online and in-store shopping, national-level modeling and investigation of the demand for online shopping with a prediction focus remain limited in the literature. Here, this paper differs from prior work and leverages two recent releases of the U.S. National Household Travel Survey (NHTS) data for 2009 and 2017 to develop machine learning (ML) models, specifically gradient boosting machine (GBM), for predicting household-level online shopping purchases. The NHTS data allow for not only conducting nationwide investigation but also at the level of households, which is more appropriate than at the individual level given the connected consumption and shopping needs of members in a household. We follow a systematic procedure for model development including employing Recursive Feature Elimination algorithm to select input variables (features) in order to reduce the risk of model overfitting and increase model explainability. Among several ML models, GBM is found to yield the best prediction accuracy. Extensive post-modeling investigation is conducted in a comparative manner between 2009 and 2017, including quantifying the importance of each input variable in predicting online shopping demand, and characterizing value-dependent relationships between demand and the input variables. In doing so, two latest advances in machine learning techniques, namely Shapley value-based feature importance and Accumulated Local Effects plots, are adopted to overcome inherent drawbacks of the popular techniques in current ML modeling. The modeling and investigation are performed at the national level, with a number of findings obtained. The models developed and insights gained can be used for online shopping-related freight demand generation and may also be considered for evaluating the potential impact of relevant policies on online shopping demand.
Utilizing a Common Model Architecture for the Simulation of On-Orbit Human Spaceflight Operations
Advances in computing and simulation technology during the 1980 s promoted the development of multiple simulations for on-orbit human spaceflight operations at the National Aeronautics and Space Administration (NASA) Johnson Space Center (JSC). By the late 1980 s, it became increasingly clear that redundancy risked leading to inefficiencies in dissemination and incorporation of models across simulations, institutional organizations, and spaceflight programs. Software development tools, such as the Trick Simulation Environment, began to emerge providing a generalized framework for the development, integration, and operation of simulations. Throughout the 1990s, engineering, operations, and training simulations began to migrate to Trick in order to take advantage of a common simulation environment. Due to forecasted reductions in funding to the Shuttle and International Space Station (ISS) programs in the early 2000s, the common framework that arose from this simulation migration was formalized as the Common Model Architecture (CMA). The CMA was established in order to further reduce redundancy, while promoting model sharing and enhancing model integrity. A minimal set of standards for data flows, functional interactions, and model organization based on Trick were defined for those features found to be most important in facilitating model interaction. Model-providing organizations within NASA JSC took ownership of models for their specific areas of expertise, and agreements were reached on common model sharing and usage. This paper describes the evolution of commonality enabled by the Trick Simulation Environment at JSC leading to the CMA, the standards defined by the CMA, a recent CMA simulation example, and potential application of the CMA to NASA s new Exploration program.
Evaluation of pressure reconstruction techniques for Model Order Reduction in incompressible convective heat transfer
This paper compares pressure reconstruction strategies in Model Order Reduction for incompressible flows with convective heat transfer. The Navier-Stokes equation are reduced along with the passive scalar transport equation for the temperature using the POD-Galerkin technique. Six different pressure reconstruction methods are evaluated, two of which are novel to the best of the authors’ knowledge. Accurate pressure reconstruction is key to avoid error buildup in when solving for the conservation of linear momentum at the reduced level. The six approaches are compared using Direct Numerical Simulations of convective heat exchange processes in a 3D Backward Facing Step with a heated cylinder. Additionally, when comparing time-averaged metrics, we observe that the reconstruction methods that approximate the reduced pressure field using techniques borrowed from full order models (mechanical analogy, pressure Poisson, and velocity supremizers) yield higher errors than the methods that seek to stabilize the reduced systems (reduced residual stabilization, artificial divergence, and Uzawa operator).
Co-optimized machine-learned manifold models for large eddy simulation of turbulent combustion
Many modeling approaches in large eddy simulation (LES) of turbulent combustion employ a projection of the thermochemical state onto a low-dimensional manifold within state space to reduce the number of transported variables and hence computational cost. Flamelet-generated manifolds (FGM) is an example of a well-established, physics-based approach, but increasingly, principal component analysis (PCA) is being used as a data-driven method for generating manifold models. For both approaches, the nonlinear relationship between the location on the predefined manifold and the outputs of interest, such as reaction rates, can be tabulated or encoded in a neural network. This work proposes a new approach for manifold modeling that extends these existing approaches. A modified neural network structure simultaneously encodes the definition of the manifold variables, the nonlinear mapping, and the subfilter closure for LES. This allows all three of these aspects of the model to be co-optimized, generating a model from any source of combustion thermochemical state data. The manifold parameterizing variables are constrained to be linear combinations of species, as in FGM and PCA-based models, to aid in interpretability and implementation. For LES, subfilter variances of the manifold variables are also included as inputs. Two types of a priori analysis are performed to evaluate the new approach. In the first, the model is trained on data from one-dimensional premixed flames. In this case, the approach recovers the behavior of flamelet-based manifold approaches, and in fact slightly improves performance by identifying an optimized progress variable. The approach is also applied to data from direct numerical simulations of spherical ignition kernels in isotropic turbulence. For any specified manifold dimensionality, the new approach provides substantially lower prediction errors than a PCA-based model developed from the same data set. Additionally, the LES formulation of the new approach can provide accurate predictions for filtered reaction rates across a variety of filter widths.
Reduction Of Sizes Of Semi-Markov Reliability Models
Trimming technique reduces computational effort by order of magnitude while introducing negligible error. Error bound depends on only three parameters from semi-Markov model: maximum sum of rates for failure transitions leaving any state, maximum average holding time for recovery-mode state, and operating time for system. Error bound computed before any model generated, enabling modeler to decide immediately whether or not model can be trimmed. Trimming procedure specified by precise and easy description, making it easy to include trimming procedure in program generating mathematical models for use in assessing reliability. Typical application of technique in design of digital control systems required to be extremely reliable. In addition to aerospace applications, fault-tolerant design has growing importance in wide range of industrial applications.
The Impact of Satellite Sea Surface Salinity for Prediction of the Coupled Indo-Pacific System
We assess the impact of satellite sea surface salinity (SSS) observations on seasonal to interannual variability of tropical Indo-Pacific Ocean dynamics as well as on dynamical ENSO forecasts. Our coupled model is composed of a primitive equation ocean model for the tropical Indo-Pacific region that is coupled with the global SPEEDY atmospheric model (Molteni, 2003). The Ensemble Reduced Order Kalman Filter is used to assimilate observations to constrain dynamics and thermodynamics for initialization of the coupled model. The baseline experiment assimilates satellite sea level, SST, and in situ subsurface temperature and salinity observations. This baseline is then compared with experiments that additionally assimilate Aquarius (version 4.0) and SMAP (version 2.0) SSS. Twelve-month forecasts are initialized for each month from Sep. 2011 to Dec. 2016. We find that including satellite SSS significantly improves NINO 3.4 sea surface temperature anomaly validation after 1 out to 12 month forecast lead times. For initialization of the coupled forecast, the positive impact of SSS assimilation is brought about by surface freshening near the eastern edge of the western Pacific warm pool and density changes that lead to shallower mixed layer between 10 degrees South latitude-5 degrees North latitude. SST differences at initialization force wide-spread downwelling favorable curl over most of the tropical Pacific. Over an average forecast, SST remains warmer with SSS assimilation at the eastern edge of the warm pool. This warm SST propagates into the eastern Pacific and drags westerly wind anomalies eastward into the NINO 3.4 region. In addition, salting near the ITCZ (Intertropical Convergence Zone) leads to a deepening of the mixed layer and thermocline near 8 degrees North latitude. These patterns together lead to a funneling effect that provides the background state to amplify equatorial Kelvin waves. We show that the downwelling Kelvin waves are amplified by assimilating satellite SSS and lead to significantly improved forecasts particularly for the 2015 El Nino.
The Impact of Satellite Sea Surface Salinity for Prediction of the Coupled Indo-Pacific System
We assess the impact of satellite sea surface salinity (SSS) observations on seasonal to interannual variability of tropical Indo-Pacific Ocean dynamics as well as on dynamical ENSO forecasts. Our coupled model is composed of a primitive equation ocean model for the tropical Indo-Pacific region that is coupled with the global SPEEDY atmospheric model (Molteni, 2003). The Ensemble Reduced Order Kalman Filter is used to assimilate observations to constrain dynamics and thermodynamics for initialization of the coupled model. The baseline experiment assimilates satellite sea level, SST, and in situ subsurface temperature and salinity observations. This baseline is then compared with experiments that additionally assimilate Aquarius (version 4.0) and SMAP (version 2.0) SSS. Twelve-month forecasts are initialized for each month from Sep. 2011 to Dec. 2016. We find that including satellite SSS significantly improves NINO3.4 sea surface temperature anomaly validation after 1 out to 12 month forecast lead times. For initialization of the coupled forecast, the positive impact of SSS assimilation is brought about by surface freshening near the eastern edge of the western Pacific warm pool and density changes that lead to shallower mixed layer between 10S-5N. SST differences at initialization force wide-spread downwelling favorable curl over most of the tropical Pacific. Over an average forecast, SST remains warmer with SSS assimilation at the eastern edge of the warm pool. This warm SST propagates into the eastern Pacific and drags westerly wind anomalies eastward into the NINO3.4 region. In addition, salting near the ITCZ leads to a deepening of the mixed layer and thermocline near 8N. These patterns together lead to a funneling effect that provides the background state to amplify equatorial Kelvin waves. We show that the downwelling Kelvin waves are amplified by assimilating satellite SSS and lead to significantly improved forecasts particularly for the 2015 El Nino.
A Transdisciplinary Review of Deep Learning Research and Its Relevance for Water Resources Scientists
Deep learning (DL), a new generation of artificial neural network research, has transformed industries, daily lives, and various scientific disciplines in recent years. DL represents significant progress in the ability of neural networks to automatically engineer problem-relevant features and capture highly complex data distributions. I argue that DL can help address several major new and old challenges facing research in water sciences such as interdisciplinarity, data discoverability, hydrologic scaling, equifinality, and needs for parameter regionalization. This review paper is intended to provide water resources scientists and hydrologists in particular with a simple technical overview, transdisciplinary progress update, and a source of inspiration about the relevance of DL to water. The review reveals that various physical and geoscientific disciplines have utilized DL to address data challenges, improve efficiency, and gain scientific insights. DL is especially suited for information extraction from image-like data and sequential data. Techniques and experiences presented in other disciplines are of high relevance to water research. Meanwhile, less noticed is that DL may also serve as a scientific exploratory tool. A new area termed AI neuroscience , where scientists interpret the decision process of deep networks and derive insights, has been born. This budding subdiscipline has demonstrated methods including correlation-based analysis, inversion of network-extracted features, reduced-order approximations by interpretable models, and attribution of network decisions to inputs. Moreover, DL can also use data to condition neurons that mimic problem-specific fundamental organizing units, thus revealing emergent behaviors of these units. Vast opportunities exist for DL to propel advances in water sciences.
Self-Secure Inverters Against Malicious Setpoints
The next generation of grid-interactive inverters brings a communication feature that allows data sharing from utility supervisory controllers and smart devices that are connected to the same network. This feature enhances the control capabilities of grid-interactive inverters to provide services beyond active power injection. However, communication networks entail more vulnerable surfaces to malicious attacks that may result in modifying active and reactive power setpoints and causing weak-grid conditions or abnormal inverter operation. In this paper, steady-state and the dynamic behavior of the inverter for the incoming setpoints are analyzed to detect false data injection attacks and provide device-level security. The steady-state behavior of the inverter in the operating region is determined from the grid parameters such as the grid voltage and the grid impedance. These estimations are accomplished by the proposed self-security technique through a low-frequency signal injection-based approach combined with the recursive least square method. Moreover, a reduced fourth-order inverter model is used as the dynamic reference model, and grid parameters as well as the incoming setpoints are implemented to the reference model to verify whether the dynamic behavior of the inverter is inside the permissible region of operation. The validity and performance of the proposed method are verified experimentally through Allen-Bradley Powerflex 755 three-phase inverter and a 12 kW NHR 9410 regenerative power grid emulator. The results show that the self-secure smart-inverter is able to accept or reject the incoming commands and thus is protected from malicious cyber-physical attacks.
Computationally Efficient Control Co-Design Optimization Framework with Mixed-Fidelity Fluid and Structure Analysis: OpenTurbineCoDe (OTCD) - ARPA-E ATLANTIS Phase I Project (Final Technical Report)
The goal of the Department of Energy (DOE) Advanced Research Projects Agency-Energy (ARPA-E) Aerodynamic Turbines Lighter and Afloat with Nautical Technologies and Integrated Servo-control (ATLANTIS) Program is to develop new technologies for floating offshore wind turbines, or wind farms, using the discipline of control co-design (CCD.) In this context, our goal is to develop a computationally efficient optimization framework for design of floating offshore wind turbines. Our specific aim is to utilize high-fidelity structural, aerodynamic, aero-structural tools, and to derive control-oriented reduced- or low-order models directly from the high-fidelity tools. We proposed a mixed-fidelity modeling approach which means that we are also using low- and mid-fidelity tools when necessary. This research is conducted by a multidisciplinary team consisting of Rutgers University, University of Michigan, Brigham Young University, and the National Renewable Energy Laboratory (NREL). The computational framework, called OpenTurbineCoDe, is designed to integrate, where possible, traditional structural, aerodynamic, aeroelastic models (e.g., OpenFAST) and advanced control algorithms with higher fidelity simulation tools including Reynolds-averaged Navier–Stokes (RANS) solvers, and three-dimensional structural finite element solvers. All the high-fidelity tools used in this research provide numerically exact gradients to facilitate both efficient optimization and local linearization for control implementation.
Precision pointing compensation for DSN antennas with optical distance measuring sensors
The pointing control loops of Deep Space Network (DSN) antennas do not account for unmodeled deflections of the primary and secondary reflectors. As a result, structural distortions due to unpredictable environmental loads can result in uncompensated boresight shifts which degrade pointing accuracy. The design proposed here can provide real-time bias commands to the pointing control system to compensate for environmental effects on pointing performance. The bias commands can be computed in real time from optically measured deflections at a number of points on the primary and secondary reflectors. Computer simulations with a reduced-order finite-element model of a DSN antenna validate the concept and lead to a proposed design by which a ten-to-one reduction in pointing uncertainty can be achieved under nominal uncertainty conditions.