Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Reduce 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.

At least 649 records · Page 36
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

NRAP-Open-IAM Analytical Reservoir Model: Development and Testing

Geological carbon sequestration (GCS) is a key technology for reducing global carbon dioxide (CO 2 ) emissions. Over the last decade, the U.S. Department of Energy has invested in understanding the science base, developing practical implementation methods, and demonstrating secure GCS technologies to mitigate the environmental impacts associated with the atmospheric release of CO 2 . As part of the National Risk Assessment Partnership, a systems-level risk assessment tool, called the NRAP-Open-IAM, has been developed to conduct risk assessment and enable safe operations at a GCS site. The current NRAP-Open-IAM contains a simple reservoir model component that calculates the evolution of CO 2 saturation and fluid pressure in a storage reservoir during CO 2 injection operations. This report presents the development and testing of a new analytical reservoir reduced-order model (ROM), which is extended from an existing semi-analytical model for estimation of CO 2 and brine leakage along legacy wells, and enhances the capability of the NRAP-Open-IAM to simulate more types of reservoir conditions. The developed model is validated against three reference studies, and the results indicate that the new ROM predicts the behavior of the two-phase fluids (brine and injected CO 2 ) well and is applicable to different reservoir simulation boundary conditions (i.e., constant pressure boundary and infinite-acting boundary) without a priori user specification of the boundary type. Sensitivity analysis for a set of model parameters is performed using 4,000 synthetic cases prepared via a fully automated process and using machine-learning-based feature selection. The stochastic analysis identifies gravitational number (i.e., ratio of gravitational forces to viscous force) and distance between the injection well and observation location as the most impactful parameters for matching the pressure and CO 2 saturation, respectively, between the numerical simulations and the ROM. This report details the possible ROM uncertainties and serves as a guide for users to understand the use and limitations of this ROM. The code implementation of the model will be released as a module within the NRAP-Open-IAM.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1986

Adaptive control robustness in flexible aircraft application

This paper presents the application of a model reference adaptive control (MRAC) algorithm to flexible aircraft flight control. The algorithm is an adaptive version of the Command Generator Tracking (CGT) control technique. This technique forces a dynamic system to follow a reduced-order model, allowing it to cope with the problem of unmodeled dynamics. The studies were made via simulation, using for the plant an aircraft dynamic model similar to the B1 bomber. This model is of a large aircraft with a reasonable amount of structural flexibility. In particular, flight configurations were analyzed where the influence of the flexible modes make it difficult to control the aircraft. The results indicate that the algorithm has good robustness properties vis-a-vis unmodeled dynamics and can force the flexible aircraft to follow rigid body responses.

Sanchez, E. N.↗
NASA NTRS1987

Model reduction for discrete bilinear systems

A model reduction method for discrete bilinear systems is developed which matches q sets of Volterra and covariance parameters. These parameters are shown to represent both deterministic and stochastic attributes of the discrete bilinear system. A reduced order model which matches these q sets of parameters is defined to be a q-Volterra covariance equivalent realization (q-Volterra COVER). An algorithm is presented which constructs a class of q-Volterra COVERs parameterized by solutions to a Hermitian, quadratic, matrix equation. The algorithm is applied to a bilinear model of a robot manipulator.

King, A. M.↗
NASA NTRS2019

Liftoff and Transition Database Generation for Launch Vehicles Using Data-Fusion-Based Modeling

A data fusion technique for merging multiple data sources with differing fidelity and resolution was developed to support the production of aerodynamic line load databases for the Liftoff and Transition (LOT) flight phase of the Space Launch System (SLS). The technique uses a reduced order model based on a high-fidelity line load data set from Computational Fluid Dynamics (CFD) to predict solutions for a much larger solution space. Even higher-fidelity force and moment information (from wind-tunnel tests) is then used to adjust the model. The adjustment uses constrained optimization through the method of Lagrange multipliers in order to minimize the deviation of the line load distribution from the spatially-dense CFD solution, while ensuring that the integrated force and moment values match those observed in physical wind tunnel measurements. Though the wind-tunnel data are operationally-dense (available at many flow conditions), they are spatially coarse (as only the overall forces and moments are available). Conversely, CFD for such complex configurations is expensive, and thus operationally sparse. Data fusion techniques are necessary to make the most efficient use of available information, delivering accurate results within time and resource constraints.

Wignall, T. J.↗
DOE OSTI2022

Energetically consistent model reduction for metriplectic systems

The metriplectic formalism is useful for describing complete dynamical systems which conserve energy and produce entropy. This creates challenges for model reduction, as the elimination of high-frequency information will generally not preserve the metriplectic structure which governs long-term stability of the system. Based on proper orthogonal decomposition, a provably convergent metriplectic reduced-order model is formulated which is guaranteed to maintain the algebraic structure necessary for energy conservation and entropy formation. Further, numerical results on benchmark problems show that the proposed method is remarkably stable, leading to improved accuracy over long time scales at a moderate increase in cost over naive methods.

42 ENGINEERING↗
NASA NTRS1989

Component mode synthesis for model order reduction of nonclassically damped systems

Based on a component mode synthesis approach, a model-order-reduction method for linear structures with arbitrary linear damping has been developed. Projection matrices are introduced to make the method applicable to systems having rigid-body freedom. To test the method, eigenvalues of a reduced-order model of a free-free beam with nonproportional damping were compared to exact eigenvalues and to eigenvalues obtained using two other model-reduction strategies. The present model-reduction strategy proved to be decidedly superior.

Craig, Roy R., Jr.↗
NASA NTRS1994

Effects of Mistuning on the Forced Response of Turbomachinery Rotors

The basic purpose of this research has been to develop a fundamental understanding of the effects of blade-to-blade dissimilarities, or mistuning, on the dynamics of nearly cyclic bladed-disk assemblies. This topic is of importance as mistuning has been shown to increase the forced response amplitudes of some blades significantly, and even to lead to blade failure. Furthermore, the current trend toward high performance propulsion turbomachinery designed for finite service life demands an accurate prediction of system performance and dynamics at the design stage. This objective has been achieved by carrying out the following tasks. First, the investigation of the free and forced responses of representative, yet sufficiently simple blade assembly models that capture all the important characteristics of typical turbomachinery rotors. Second, the development of computational methods that predict the effects of mistuning in a systematic and reliable way, along with the development of a systematic reduced-order modeling procedure for mistuned bladed disks. Third, the application of these findings and the tools developed to an industrial rotor, namely the first stage of turbine blades of the oxidizer turbopump in the space shuttle main rocket engine (SSME). The research supported by NASA has led to the development of a coherent theory for mistuned blade assemblies. The further implementation of these computational tools into the forced response prediction system currently under development in the Structural Dynamics Branch-FREPS-would enable the designer and the analyst to: (1) identify types of blade assemblies highly sensitive to mistuning in various frequency and other parameter ranges and (2) characterize mistuning effects on their forced response by predicting true response amplitudes and fatigue life estimates.

Pierre, Christophe↗
NASA NTRS1994

Effects of Mistuning on the Forced Response of Turbomachinery Rotors

The basic purpose of this research has been to develop a fundamental understanding of the effects of blade-to-blade dissimilarities, or mistuning, on the dynamics of nearly cyclic bladed-disk assemblies. This topic is of importance as mistuning has been shown to increase the forced response amplitudes of some blades significantly, and even to lead to blade failure. Furthermore, the current trend toward high performance propulsion turbomachinery designed for finite service life demands an accurate prediction of system performance and dynamics at the design stage. This objective has been achieved by carrying out the following tasks. First, the investigation of the free and forced responses of representative, yet sufficiently simple blade assembly models that capture all the important characteristics of typical turbomachinery rotors. Second, the development of computational methods that predict the effects of mistuning in a systematic and reliable way, along with the development, of a systematic reduced-order modeling procedure for mistuned bladed disks. Third, the application of these findings and the tools developed to an industrial rotor, namely the first stage of turbine blades of the oxidizer turbopump in the space shuttle main rocket engine (SSME). The research supported by NASA has led to the development of a coherent theory for mistuned blade assemblies. The further implementation of these computational tools into the forced response prediction system currently under development in the Structural Dynamics Branch Forced Response Prediction System (FREPS) would enable the designer and the analyst (1) to identify types of blade assemblies highly sensitive to mistuning in various frequency and other parameter ranges and (2), to characterize mistuning effects on their forced response by predicting true response amplitudes and fatigue life estimates.

Stefko, George L.↗
NASA NTRSDate not supplied

DeepONet-Assisted Optimization of Surface Topography for Transition Delay in a Mach 4.5 Boundary Layer

We use deep learning, an ensemble variational technique (EnVar), and direct numerical simulations(DNS) to design an optimal topography for a two-dimensional roughness element that delays the on-set of laminar-turbulent transition in a Mach 4.5 flat-plate boundary layer. Deep operator networks (DeepONets), which have the known ability to learn complex nonlinear operators within dynamical systems, are used for machine learning. For the baseline configuration of a smooth flat plate, the second-mode waves at the DNS inflow cause a quick nonlinear breakdown of the high-speed boundary layer within the computational domain. Results reported in the present study validate the ability of DeepONets to model the transition delay via a given topography of the roughness element. The computing cost to optimize the rough-ness element for minimal skin-friction drag is substantially lowered by the DeepONets-based reduced-order model. In comparison to the baseline method of EnVar optimization based on DNS alone, the DeepONets-based EnVar optimizer is able to delay transition past the outflow boundary of the computational domain while utilizing almost 5–6 times fewer DNS.

Machine Learning↗
NASA NTRS1991

The process of control design for the NASA Langley Minimast structure

The design process used to select sensors for feedback and performance weights on the Minimast facility is discussed. Initially, a series of controllers are synthesized using H2 optimal control techniques for the given structural model, a variety of sensor locations and performance criteria to determine the best displacement sensor and/or accelerometers to be used for feedback. Upon selection of the sensors, controllers are formulated to determine the affect of using a reduced-order model of the Minimast structure instead of the higher-order structural analysis model for control design and the relationship between the actuator torque level and the closed-loop performance. Based on this information, controllers are designed using micro-synthesis techniques and implemented on the Minimast structure. Results of the implementation of these controllers on the Minimast experimental facility are presented.

Balas, Gary J.↗
DOE OSTI2020

Characterizing magnetized plasmas with dynamic mode decomposition

Accurate and efficient plasma models are essential to understand and control experimental devices. Existing magnetohydrodynamic or kinetic models are nonlinear and computationally intensive and can be difficult to interpret, while often only approximating the true dynamics. In this work, data-driven techniques recently developed in the field of fluid dynamics are leveraged to develop interpretable reduced-order models of plasmas that strike a balance between accuracy and efficiency. In particular, dynamic mode decomposition (DMD) is used to extract spatio-temporal magnetic coherent structures from the experimental and simulation datasets of the helicity injected torus with steady inductive (HIT-SI) experiment. Three-dimensional magnetic surface probes from the HIT-SI experiment are analyzed, along with companion simulations with synthetic internal magnetic probes. A number of leading variants of the DMD algorithm are compared, including the sparsity-promoting and optimized DMD. Optimized DMD results in the highest overall prediction accuracy, while sparsity-promoting DMD yields physically interpretable models that avoid overfitting. These DMD algorithms uncover several coherent magnetic modes that provide new physical insights into the inner plasma structure. These modes were subsequently used to discover a previously unobserved three-dimensional structure in the simulation, rotating at the second injector harmonic. Finally, using data from probes at experimentally accessible locations, DMD identifies a resistive kink mode, a ubiquitous instability seen in magnetized plasmas.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1989

Kalman estimation for SEDS measurements

The first on-orbit experiment of the Small Expendable Deployer System (SEDS) for tethered satellites will collect telemetry data for tethered length, rate of deployment, and tether tension. The post-flight analysis will use this data to reconstruct the deployment history and determine payload position and tether shape. Two Kalman estimator algorithms were written, and output using simulated measurement data was compared. Both estimators exhibited the same estimated state histories, indicating that numerical instability in the traditional algorithm was not the cause of filter divergence. Estimation of acceleration biases was added, which reduced the error but did not correct the divergence. An add-a-bead estimator that adds lumped masses as the tether is deployed was written, which provides a state model that matches the BEADSIM simulation providing the true measurements and states. This twenty-one bead estimator produced state histories similar to those of the two-bead estimator, indicating that the filter divergence was not caused by a reduced-order model. The noise models used to date are relatively simple and may be the source of estimator divergence. The investigation of colored noise models, cross-correlated measurement and process covariances, and noise-adaptive filter techniques is recommended.

Carrington, Connie K.↗
NASA NTRS1990

Model reduction by weighted Component Cost Analysis

Component Cost Analysis considers any given system driven by a white noise process as an interconnection of different components, and assigns a metric called 'component cost' to each component. These component costs measure the contribution of each component to a predefined quadratic cost function. A reduced-order model of the given system may be obtained by deleting those components that have the smallest component costs. The theory of Component Cost Analysis is extended to include finite-bandwidth colored noises. The results also apply when actuators have dynamics of their own. Closed-form analytical expressions of component costs are also derived for a mechanical system described by its modal data. This is very useful to compute the modal costs of very high order systems. A numerical example for MINIMAST system is presented.

Kim, Jae H.↗
DOE OSTI2020

Probabilistic neural networks for fluid flow surrogate modeling and data recovery

We consider the use of probabilistic neural networks for fluid flow surrogate modeling and data recovery. This framework is constructed by assuming that the target variables are sampled from a Gaussian distribution conditioned on the inputs. Consequently, the overall formulation sets up a procedure to predict the hyperparameters of this distribution which are then used to compute an objective function given training data. We demonstrate that this framework has the ability to provide for prediction confidence intervals based on the assumption of a probabilistic posterior, given an appropriate model architecture and adequate training data. The applicability of the present framework to cases with noisy measurements and limited observations is also assessed. To demonstrate the capabilities of this framework, we consider canonical regression problems of fluid dynamics from the viewpoint of reduced-order modeling and spatial data recovery for four canonical data sets. The examples considered in this study arise from (i) the shallow-water equations, (ii) a two-dimensional cylinder flow, (iii) the wake of a NACA0012 airfoil with a Gurney flap, and (iv) the NOAA sea surface temperature data set. Furthermore, the present results indicate that the probabilistic neural network not only produces a machine-learning-based fluid flow surrogate model but also systematically quantifies the uncertainty therein to assist with model interpretability.

42 ENGINEERING↗
DOE OSTI2021

Effectiveness and nonlinear characterization of vibro-impact energy harvesting absorbers in controlling base-excited systems

In this study, model derivations are carried out of a dynamical system under base excitations with a piezoelectric energy harvesting absorber as the tuned-mass-damper. Additionally, amplitude stoppers are included to the absorber in order to create a broadband resonant response, increasing the window of operational use for energy harvesting and system's control. This study is unique in the fact that the energy harvester is coupled to the source of its excitation. A nonlinear reduced-order model is developed using Euler–Lagrange principle and the Galerkin method to accurately estimate the energy harvesting absorber's displacement, harvested power, and the oscillating response of the primary structure. The nonlinear interaction of the energy harvesting absorber and the amplitude stoppers are the focus of this study, where an in-depth investigation of bifurcation points of the primary structure and energy harvesting absorber responses is performed. Due to a transfer of energy between the primary structure and the absorber, it is shown that a soft stopper with stiffness $5 \times {10^3}\,{\text{N}}\;{{\text{m}}^{ - 1}}\,$ has great control of the primary structure with 60% of the uncontrolled amplitude being reduced, as well as an increase of the harvested energy. Medium stoppers with small initial gaps size and hard stoppers do not control the primary structure and show a decrease in the energy harvesting capabilities due to the activation of the nonlinear contact-impact interactions. Finally, these stoppers also generate aperiodic regions due to the possible presence of grazing bifurcations.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Dynamic probabilistic risk assessment and game theory for cyber security risk analysis in nuclear power plants

Nuclear Power Plants and energy systems have become more prone to cyber-attacks with their digitalization and the increased use of smart equipment. Hence, it is important to quantify the risk associated with cyber-attacks in such systems. Dynamic Probabilistic Risk Assessment which involves studying the evolution of a system due to random events and operator and attacker actions during a cyber-attack by employing a physics-based model of the system is a suitable framework to quantify cybersecurity risk in nuclear power plants. In addition to the plant dynamics, it is also important to model the strategies of the attackers and plant operators for an effective cybersecurity risk assessment. Game theory provides a set of necessary tools to model such strategic interactions. In this research, a framework that integrates dynamic probabilistic risk assessment with game theory for cybersecurity risk analysis in nuclear power plants is presented. The mathematical formulation is derived based on the theory of continuous event trees. We propose a game theory based action model, that utilizes physics-based rewards to define the strategies of attackers and operators at every decision epoch. As a case study, the risk associated with cyber-attacks on the digital components in the secondary side of a pressurized water reactor is studied using a reduced order model. A set of attacker actions and a set of operator actions are defined for the system. The operator and attacker interactions were modelled using simultaneous game, their action policies were computed using the concept of mixed strategy Nash equilibrium and the evolution of the system was studied.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

The Integrated Virtual Blast Furnace: Enabling Physics-Based Operational Guidance

As part of a DOE-supported research effort, Purdue University Northwest researchers are collaborating with Oak Ridge National Laboratory and United States Steel Corporation to develop a tool to provide blast furnace operators and engineers with process performance insight comparable to high-fidelity computational fluid dynamics modeling, accelerated to provide “what-if” scenarios at near-real-time speed. This is accomplished by pre-simulating a baseline case and a range of potential operating scenarios to establish how the furnace responds to changing inputs, then training a neural network-based Reduced Order Model to accelerate the speed at which predictions of key parameters can be generated.

Okosun, Tyamo↗
NASA NTRS2017

Computational Results for the KTH-NASA Wind-Tunnel Model Used for Acquisition of Transonic Nonlinear Aeroelastic Data

A status report is provided on the collaboration between the Royal Institute of Technology (KTH) in Sweden and the NASA Langley Research Center regarding the aeroelastic analyses of a full-span fighter configuration wind-tunnel model. This wind-tunnel model was tested in the Transonic Dynamics Tunnel (TDT) in the summer of 2016. Large amounts of data were acquired including steady/unsteady pressures, accelerations, strains, and measured dynamic deformations. The aeroelastic analyses presented include linear aeroelastic analyses, CFD steady analyses, and analyses using CFD-based reduced-order models (ROMs).

Silva, Walter A.↗