Reducing Model Biases in Aerosols and Clouds through Recent Airborne Field Campaigns
Explore the source record for details and available documents.
Allison Collow↗
SEARCH · Search NASA
Results for “Reduced models”
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.
Design of multivariable feedback control systems via spectral assignment using reduced-order models and reduced-order observers
The feasibility of using reduced order models and reduced order observers with eigenvalue/eigenvector assignment procedures is investigated. A review of spectral assignment synthesis procedures is presented. Then, a reduced order model which retains essential system characteristics is formulated. A constant state feedback matrix which assigns desired closed loop eigenvalues and approximates specified closed loop eigenvectors is calculated for the reduced order model. It is shown that the eigenvalue and eigenvector assignments made in the reduced order system are retained when the feedback matrix is implemented about the full order system. In addition, those modes and associated eigenvectors which are not included in the reduced order model remain unchanged in the closed loop full order system. The full state feedback design is then implemented by using a reduced order observer. It is shown that the eigenvalue and eigenvector assignments of the closed loop full order system rmain unchanged when a reduced order observer is used. The design procedure is illustrated by an actual design problem.
Mielke, R. R.↗
Design of multivariable feedback control systems via spectral assignment using reduced-order models and reduced-order observers
The feasibility of using reduced order models and reduced order observers with eigenvalue/eigenvector assignment procedures is investigated. A review of spectral assignment synthesis procedures is presented. Then, a reduced order model which retains essential system characteristics is formulated. A constant state feedback matrix which assigns desired closed loop eigenvalues and approximates specified closed loop eigenvectors is calculated for the reduced order model. It is shown that the eigenvalue and eigenvector assignments made in the reduced order system are retained when the feedback matrix is implemented about the full order system. In addition, those modes and associated eigenvectors which are not included in the reduced order model remain unchanged in the closed loop full order system. The fulll state feedback design is then implemented by using a reduced order observer. It is shown that the eigenvalue and eigenvector assignments of the closed loop full order system remain unchanged when a reduced order observer is used. The design procedure is illustrated by an actual design problem.
Mielke, R. R.↗
Investigation on the Practicality of Developing Reduced Thermal Models
Throughout the spacecraft design and development process, detailed instrument thermal models are created to simulate their on-orbit behavior and to ensure that they do not exceed any thermal limits. These detailed models, while generating highly accurate predictions, can sometimes lead to long simulation run times, especially when integrated with a spacecraft observatory model. Therefore, reduced models containing less detail are typically produced in tandem with the detailed models so that results may be more readily available, albeit less accurate. In the current study, both reduced and detailed instrument models are integrated with their associated spacecraft bus models to examine the impact of instrument model reduction on run time and accuracy. Preexisting instrument bus thermal model pairs from several projects were used to determine trends between detailed and reduced thermal models; namely, the Mirror Optical Bench (MOB) on the Gravity and Extreme Magnetism Small Explorer (GEMS) spacecraft, Advanced Topography Laser Altimeter System (ATLAS) on the Ice, Cloud, and Elevation Satellite 2 (ICESat-2), and the Neutral Mass Spectrometer (NMS) on the Lunar Atmosphere and Dust Environment Explorer (LADEE). Hot and cold cases were run for each model to capture the behavior of the models at both thermal extremes. It was found that, though decreasing the number of nodes from a detailed to reduced model brought about a reduction in the run-time, a large time savings was not observed, nor was it a linear relationship between the percentage of nodes reduced and time saved. However, significant losses in accuracy were observed with greater model reduction. It was found that while reduced models are useful in decreasing run time, there exists a threshold of reduction where, once exceeded, the loss in accuracy outweighs the benefit from reduced model runtime.
0000↗
Adaptive tracking for complex systems using reduced-order models
Reduced-order models are considered in the context of parameter adaptive controllers for tracking workspace trajectories. A dual-arm manipulation task is used to illustrate the methodology and provide simulation results. A parameter adaptive controller is designed to track the desired position trajectory of a payload using a four-parameter model instead of a full-order, nine-parameter model. Several simulations with different payload-to-arm mass ratios are used to illustrate the capabilities of the reduced-order model in tracking the desired trajectory.
Carignan, Craig R.↗
Adaptive tracking for complex systems using reduced-order models
Reduced-order models are considered in the context of parameter adaptive controllers for tracking workspace trajectories. A dual-arm manipulation task is used to illustrate the methodology and provide simulation results. A parameter adaptive controller is designed to track a payload trajectory using a four-parameter model instead of the full-order, nine-parameter model. Several simulations with different payload-to-arm mass ratios are used to illustrate the capabilities of the reduced-order model in tracking the desired trajectory.
Carnigan, Craig R.↗
Additive Manufacturing Model-Based Process Metrics: Reduced Order Modeling of the Laser Powder Bed Fusion Process
The multi-scale and complex process of printing additively manufactured (AM) parts can have unexpected, but predictable, build conditions that result in material microstructure variability. In this work, we describe a fully parallel reduced order modeling approach that has been developed to evaluate the evolution of AM processes, termed the AM moment measure method. This method couples the known sequence of the AM process with a physically informed nearest neighbors’ calculation to map the conditions of a part-scale build. The result is a map of the build that is derived directly from build files or in-situ process monitoring sensors. The methodology and terminology of the approach will be described, and computed build maps will be calculated and compared for various laser powder bed fusion (LPBF) builds of Ti-6Al-4V. Such comparative results develop understanding of how the sequential process actions can affect the LPBF-AM build quality and microstructure variability.
Laser Powder Bed Fusion↗
Quantifying Emergent Fluid Dynamics Using Reynolds-Interpolated Fluid Reduced-order Models
Fluid reduced-order models (ROMs) which capture the flow physics within the problem's physical domain are usually constrained in accuracy to only the parameter points, e.g. Reynolds and Mach numbers, at which reference data was provided. Interpolation-focused quantity-of-interest ROMs are often structured differently and fail to provide flow volume data with the same quality - if at all. In this paper, techniques which reside at the intersection of these two ROM schools - flow physics ROMs which can be interpolated within a parameter space of interest - are explored. Using a combination of existing and novel techniques, emergent physics are identified using a fluid ROM at parameter points which are not provided in the ROM's training data.
uncertainty quantification↗
Quantifying Emergent Fluid Dynamics Using Reynolds-Interpolated Fluid Reduced-order Models
Fluid reduced-order models (ROMs) which capture the flow physics within the problem's physical domain are usually constrained in accuracy to only the parameter points, e.g. Reynolds and Mach numbers, at which reference data was provided. Interpolation-focused quantity-of-interest ROMs are often structured differently and fail to provide flow volume data with the same quality - if at all. In this paper, techniques which reside at the intersection of these two ROM schools - flow physics ROMs which can be interpolated within a parameter space of interest - are explored. Using a combination of existing and novel techniques, emergent physics are identified using a fluid ROM at parameter points which are not provided in the ROM's training data.
uncertainty quantification↗
Dominant modes via model error
Obtaining a reduced model of a stable mechanical system with proportional damping is considered. Such systems can be conveniently represented in modal coordinates. Two popular schemes, the modal cost analysis and the balancing method, offer simple means of identifying dominant modes for retention in the reduced model. The dominance is measured via the modal costs in the case of modal cost analysis and via the singular values of the Gramian-product in the case of balancing. Though these measures do not exactly reflect the more appropriate model error, which is the H2 norm of the output-error between the full and the reduced models, they do lead to simple computations. Normally, the model error is computed after the reduced model is obtained, since it is believed that, in general, the model error cannot be easily computed a priori. The authors point out that the model error can also be calculated a priori, just as easily as the above measures. Hence, the model error itself can be used to determine the dominant modes. Moreover, the simplicity of the computations does not presume any special properties of the system, such as small damping, orthogonal symmetry, etc.
Yousuff, A.↗
Early climate on earth-reduced gas models and early climate on Mars-reduced gas and obliquity models
At high obliquity, Martian polar ground temperatures could exceed the melting point of ice for considerable periods of time (approximately 90 Earth days). Under special conditions ice itself might melt. Carbon dioxide adsorbed on the Martian regolith is not expected to buffer the seasonal pressure wave except in the unlikely event that the soil pore size is very large (50 micron). For a basaltic soil composition the maximum CO2 that could be desorbed over obliquity time scales due to thermal forces is a few millibars. At low obliquities the atmospheric pressures may drop, desorbing the soil. The only means to achieve higher CO2 pressures is to have much higher planet-wide temperatures due to some greenhouse effect, or to be at an epoch before the regolith or carbonates formed. The water ice budget between north and south polar caps was considered and summer sublimation rates imply that the ice could be exchanged between the poles during obliquity cycles. A critical factor in the polar cap water budget is the interaction of water and dust. The origin of the Martian polar laminae is probably due to variations in this interaction.
Toon, O. B.↗
Reduced order models for nonlinear aerodynamics
Reduced order models are needed for reliable, efficient and accurate prediction of aerodynamic forces to analyze fluid structure interaction problems in turbomachinery, including propfans. Here, a finite difference, time marching Navier-Stokes code is validated for unsteady airfoil motion by comparing results with those from classical potential flow. The Navier-Stokes code is then analyzed for calculation of primitive and exact estimates of eigenvalues and eigenvectors associated with fluid-airfoil interaction. A variational formulation for the Euler equations and Navier-Stokes equations will be the basis for reduction of order through an eigenvector transformation.
Mahajan, Aparajit J.↗
Stochastic Reduced Order Models with Python (SROMPy)
Stochastic Reduced Order Models with Python (SROMPy) is a software package developed to enable user-friendly utilization of the stochastic reduced order model (SROM) approach for uncertainty quantification. A SROM is a low dimensional, discrete approximation to a random quantity that enables efficient and non-intrusive stochastic computations. With SROMPy, a user can easily generate a SROM to approximate a random variable or vector described by several different types of probability distributions using the Python programming language. Once a SROM is constructed, the software can be used to propagate uncertainty through a user-defined computational model to estimate statistics of a given quantity of interest. This report is meant to introduce the SROMPy module and brie y demonstrate its capabilities. A simple example of a spring-mass system with a random input is included to illustrate the practicality of the SROM approach to uncertainty quantification and relative ease of applying it with SROMPy. The example includes a comparison with a solution obtained using classical Monte Carlo simulation, demonstrating the similarities and advantages of using the SROM approach.
Warner, James E.↗
Alternate Methods of Model Reduction to Avoid Dynamic Modal Truncation Error
Loads analysis is traditionally performed using dynamically reduced models, which provides the benefit to reduce run time. If the reduced model frequency cutoff is not chosen appropriately, the model will lack the dynamic content required to fully represent the response of the non-reduced model. Standard guidance for reduced model frequency content, provided in NASA-STD-5002, is to solve fixed base modes up to a minimum of 1.5x the model frequency content of interest and to employ static modal truncation methods such as residual vectors, the mode acceleration method, and the residual flexibility method to account for the truncated flexibility of the missing modes. Fixed base modes require the selection of a set of degrees of freedom to be constrained which, if not properly selected, may affect the accuracy of the reduced model by excluding some of the dynamic characteristic of the full model. In this case, the standard NASA guidance would be insufficient, but it may not be readily apparent that a portion of the reduced model response is missing. This error was encountered during an independent verification and validation (IV&V) effort, where it was observed that the resulting dynamic response was lower than the inline analysis. In this specific case, despite following the standard NASA model reduction guidelines in the selection of the frequency cutoff, the inline model still did not fully capture the necessary dynamic content. As part of the IV&V, an alternate reduction methodology was employed using an unconstrained mode acceleration method. The original model initially performed a constrained reduction to twice the frequency content of interest before doing a free-free run, employing the mode acceleration method to account for the truncated modes. In contrast in the IV&V, the unconstrained model reduced directly to the needed frequency content of the free-free run, avoiding any interactions between constraints and dynamic content. To verify the model, the original reduction methodology was used to generate a series of Hurty-Craig-Bampton reductions, each with a higher frequency cutoff than the previous. The results were shown to converge once the frequency cutoff increased past eight times the frequency content of interest. At the same frequency cutoff, the results of the Hurty-Craig-Bampton model converged with the results of the unconstrained mode acceleration model. This comparative study provided confidence that the results of the unconstrained modal acceleration reduced model were correct and that the Hurty-Craig-Bampton needed to increase its frequency cutoff to fully capture the dynamic response.
Erin Simmons↗
Alternate Methods of Model Reduction to Avoid Dynamic Modal Truncation Error
Loads analysis is traditionally performed using dynamically reduced models, which provides the benefit to reduce run time. If the reduced model frequency cutoff is not chosen appropriately, the model will lack the dynamic content required to fully represent the response of the non-reduced model. Standard guidance for reduced model frequency content, provided in NASA-STD-5002, is to solve fixed base modes up to a minimum of 1.5x the model frequency content of interest and to employ static modal truncation methods such as residual vectors, the mode acceleration method, and the residual flexibility method to account for the truncated flexibility of the missing modes. Fixed base modes require the selection of a set of degrees of freedom to be constrained which, if not properly selected, may affect the accuracy of the reduced model by excluding some of the dynamic characteristic of the full model. In this case, the standard NASA guidance would be insufficient, but it may not be readily apparent that a portion of the reduced model response is missing. This error was encountered during an independent verification and validation (IV&V) effort, where it was observed that the resulting dynamic response was lower than the inline analysis. In this specific case, despite following the standard NASA model reduction guidelines in the selection of the frequency cutoff, the inline model still did not fully capture the necessary dynamic content. As part of the IV&V, an alternate reduction methodology was employed using an unconstrained mode acceleration method. The original model initially performed a constrained reduction to twice the frequency content of interest before doing a free-free run, employing the mode acceleration method to account for the truncated modes. In contrast in the IV&V, the unconstrained model reduced directly to the needed frequency content of the free-free run, avoiding any interactions between constraints and dynamic content. To verify the model, the original reduction methodology was used to generate a series of Hurty-Craig-Bampton reductions, each with a higher frequency cutoff than the previous. The results were shown to converge once the frequency cutoff increased past eight times the frequency content of interest. At the same frequency cutoff, the results of the Hurty-Craig-Bampton model converged with the results of the unconstrained mode acceleration model. This comparative study provided confidence that the results of the unconstrained modal acceleration reduced model were correct and that the Hurty-Craig-Bampton needed to increase its frequency cutoff to fully capture the dynamic response.
Erin Simmons↗
Accuracy of derivatives of control performance using a reduced structural model
The sensitivity of control system performance to structural changes is calculated for a multi-span beam with direct-rate feedback vibration control. Reduced models based on the natural modes of the structure are formed and derivatives of the damping ratios of the closed-loop eigenvalues are calculated. The convergence of the derivatives of the damping ratios with increasing number of modes is shown to be slower than the convergence of the damping ratios themselves. In particular, in some cases the convergence of finite-element approximations to the derivatives is much faster than the convergence of the modal approximations. The results indicate that the use of reduced models based on natural vibration modes may be ill-advised for calculating the sensitivity of control system performance to changes in the controlled structure.
Sandridge, Chris A.↗
Uncertainty Aware Structural Topology Optimization Via a Stochastic Reduced Order Model Approach
This work presents a stochastic reduced order modeling strategy for the quantification and propagation of uncertainties in topology optimization. Uncertainty aware optimization problems can be computationally complex due to the substantial number of model evaluations that are necessary to accurately quantify and propagate uncertainties. This computational complexity is greatly magnified if a high-fidelity, physics-based numerical model is used for the topology optimization calculations. Stochastic reduced order model (SROM) methods are applied here to effectively 1) alleviate the prohibitive computational cost associated with an uncertainty aware topology optimization problem; and 2) quantify and propagate the inherent uncertainties due to design imperfections. A generic SROM framework that transforms the uncertainty aware, stochastic topology optimization problem into a deterministic optimization problem that relies only on independent calls to a deterministic numerical model is presented. This approach facilitates the use of existing optimization and modeling tools to accurately solve the uncertainty aware topology optimization problems in a fraction of the computational demand required by Monte Carlo methods. Finally, an example in structural topology optimization is presented to demonstrate the effectiveness of the proposed uncertainty aware structural topology optimization approach.
Aguilo, Miguel A.↗
Achieving High Efficiency in Reduced Order Modeling for Large Scale Polycrystal Plasticity Simulations
Reduced order models for the nonlinear response of heterogeneous microstructures typically require a construction (or training) stage to build the reduced order basis. In this manuscript, an efficient model construction strategy for the eigenstrain homogenization method (EHM) is presented. The proposed strategy relies on a parallel, element-by-element, conjugate gradient solver. Near linear scaling has been achieved with respect to the number of degrees of freedom used to resolve the microstructure. Linear scaling with respect to the number of pre-analyses required to construct the reduced order model (ROM) follows from the EHM formulation. Furthermore, a parallel implementation for fast evaluation of the constructed ROM has been developed using shared memory parallelization. It has been shown that for large microstructures with ≈ 10,000 grains, the total computational cost of evaluating the nonlinear response of a polycrystal could be reduced by approximately an order of magnitude using 32 cores with respect to serial ROM simulation. The present methodology has been verified using an additively manufactured polycrystalline microstructure of a nickel-based superalloy, Inconel 625. The capability of the developed framework to construct a ROM for such large microstructures, as well as the ability of the ROM to predict average and local quantities of interest has been demonstrated.
microscale↗