SEARCH · Engineering Papers
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 336 degrees of freedom slave finite element processing capability to analyze engine structures under severe thermomechanical loading is presented. Description of the theoretical development and demonstration of that element is presented in this volume.
The nonlinear dynamics following saturation of the parametric instabilities of a monochromatic field-aligned large-amplitude circularly polarized Alfven wave is investigated via direct numerical simulation in the case of high plasma beta and no wave dispersion. The magnetohydrodynamic (MHD) code permits nonlinear couplings in the parallel direction to the ambient magnetic field and one perpendicular direction. Compressibility is included in the form of a polytropic equation of state. Turbulent cascades develop after saturation of two coupled oblique three-wave parametric instabilities; one of which is an oblique filamentationlike instability reported earlier. Remnants of the parametric processes, as well as of the original Alfven pump wave, persist during late nonlinear times. Nearly incompressible MHD features such as spectral anisotropies appear as well.
Magnetic reconnection, a plasma process converting magnetic energy to particle kinetic energy, is often invoked to explain magnetic energy releases powering high-energy flares in astrophysical sources including pulsar wind nebulae and black hole jets. Reconnection is usually seen as the (essentially two-dimensional) nonlinear evolution of the tearing instability disrupting a thin current sheet. To test how this process operates in three dimensions, we conduct a comprehensive particle-in-cell simulation study comparing two- and three-dimensional evolution of long, thin current sheets in moderately magnetized, collisionless, relativistically hot electron–positron plasma, and find dramatic differences. We first systematically characterize this process in two dimensions, where classic, hierarchical plasmoid-chain reconnection determines energy release, and explore a wide range of initial configurations, guide magnetic field strengths and system sizes. We then show that three-dimensional (3-D) simulations of similar configurations exhibit a diversity of behaviours, including some where energy release is determined by the nonlinear relativistic drift-kink instability. Thus, 3-D current sheet evolution is not always fundamentally classical reconnection with perturbing 3-D effects but, rather, a complex interplay of multiple linear and nonlinear instabilities whose relative importance depends sensitively on the ambient plasma, minor configuration details and even stochastic events. It often yields slower but longer-lasting and ultimately greater magnetic energy release than in two dimensions. Intriguingly, non-thermal particle acceleration is astonishingly robust, depending on the upstream magnetization and guide field, but otherwise yielding similar particle energy spectra in two and three dimensions. Although the variety of underlying current sheet behaviours is interesting, the similarities in overall energy release and particle spectra may be more remarkable.
A phasic excitation-activation (PEXA) model is presented of the process of motoneuron excitation and the resultant activation and force development of a motor unit. The model input is an amount of depolarizing current (as when injected with an intracellular electrode), and the model output is muscle force. The model includes dynamics and nonlinearities similar to phenomena discovered experimentally by others: the firing rate response of motoneurons to steps of depolarizing current and the catch-like enhancement of force produced by overlapping motor neuron action potentials. The parameter values used in this model are derived from experimentally measured data and are expressed in physical units. Model predictions extend to published data beyond those used in generating the model parameter values.
Here we introduce a denoising diffusion algorithm to discover microstructures with nonlinear fine-tuned properties. Denoising diffusion probabilistic models are generative models that use diffusion-based dynamics to gradually denoise images and generate realistic synthetic samples. By learning the reverse of a Markov diffusion process, we design an artificial intelligence to efficiently manipulate the topology of microstructures to generate a massive number of prototypes that exhibit constitutive responses sufficiently close to designated nonlinear constitutive behaviors. To identify the subset of microcstructures with sufficiently precise fine-tuned properties, a convolutional neural network surrogate is trained to replace high-fidelity finite element simulations to filter out prototypes outside the admissible range. Results of this study indicate that the denoising diffusion process is capable of creating microstructures of fine-tuned nonlinear material properties within the latent space of the training data. More importantly, this denoising diffusion algorithm can be easily extended to incorporate additional topological and geometric modifications by introducing high-dimensional structures embedded in the latent space. Numerical experiments are conducted on the open-source mechanical MNIST data set (Lejeune, 2020). Consequently, this algorithm is not only capable of performing inverse design of nonlinear effective media, but also learns the nonlinear structure–property map to quantitatively understand the multiscale interplay among the geometry, topology, and their effective macroscopic properties.
High-energy-density physics is the field of physics concerned with studying matter at extremely high temperatures and densities. Such conditions produce highly nonlinear plasmas, in which several phenomena that can normally be treated independently of one another become strongly coupled. The study of these plasmas is important for our understanding of astrophysics, nuclear fusion and fundamental physics—however, the nonlinearities and strong couplings present in these extreme physical systems makes them very difficult to understand theoretically or to optimize experimentally. Here we argue that machine learning models and data-driven methods are in the process of reshaping our exploration of these extreme systems that have hitherto proved far too nonlinear for human researchers. Furthermore, from a fundamental perspective, our understanding can be improved by the way in which machine learning models can rapidly discover complex interactions in large datasets. From a practical point of view, the newest generation of extreme physics facilities can perform experiments multiple times a second (as opposed to approximately daily), thus moving away from human-based control towards automatic control based on real-time interpretation of diagnostic data and updates of the physics model. To make the most of these emerging opportunities, we suggest proposals for the community in terms of research design, training, best practice and support for synthetic diagnostics and data analysis.
The remarkable mechanical properties and thermal stability of fibers fabricated from liquid crystalline polymers (LCPs) have led to the use of these materials in structural applications where weight savings are critical. Advances in processing of LCPs could permit the incorporation of these polymers into other than uniaxial designs and extend their utility into new areas such as nonlinear optical devices. However, the unique feature of LCPs (intrinsic orientation order) is itself problematic, and current understanding of processing with control of orientation falls short of allowing manipulation of macroscopic orientation (except for the case of uniaxial fibers). The current and desirable characteristics of LCPs are reviewed and specific problems are identified along with issues that must be addressed so that advances in the use of these unique polymers can be expedited.
In nonlinear optics, significant effort is concentrated on improving the strength and efficiency of interactions; however, experimentally investigating nonlinear materials is a complex, time-consuming, and costly investment. Moreover, it is often challenging to isolate, study, and optimize material parameters in an experiment due to complexities in the growth process. Recently, epsilon-near-zero materials have received a great deal of attention as promising nonlinear optical materials, but like many up-and-coming materials, the ability to explore and optimize their properties has been challenging. Here, we establish a framework to rapidly evaluate the performance of nonlinear epsilon-near-zero materials for both inter- and intraband effects in silico, requiring only an energy-momentum (E-k) diagram, linear optical properties, and experimental conditions. Measured nonlinear reflection and transmission in gallium-doped zinc oxide films are compared to the numerical framework for both intra- and interband excitation to verify accuracy across wavelength and irradiance while two figures of merit (FoMs) are introduced to quickly evaluate the performance of films without a full numerical framework. This capability is used to predict the performance of highly doped gallium nitride, cadmium oxide, zinc oxide, and indium tin oxide films, and efficient intra- and interband operation conditions are identified. Through this numerical framework and the FoMs, the exploration of unstudied epsilon-near-zero materials is enabled without the need for a nonlinear experiment, thereby accelerating the search for more efficient nonlinear materials and excitation conditions.
Capabilities and expertise related to the development of links between nondestructive evaluation (NDE) and finite element analysis (FEA) at Glenn Research Center (GRC) are demonstrated. Current tools to analyze data produced by computed tomography (CT) scans are exercised to help assess the damage state in high temperature structural composite materials. A utility translator was written to convert velocity (an image processing software) STL data file to a suitable CAD-FEA type file. Finite element analyses are carried out with MARC, a commercial nonlinear finite element code, and the analytical results are discussed. Modeling was established by building MSC/Patran (a pre and post processing finite element package) generated model and comparing it to a model generated by Velocity in conjunction with MSC/Patran Graphics. Modeling issues and results are discussed in this paper. The entire process that outlines the tie between the data extracted via NDE and the finite element modeling and analysis is fully described.
Aerosol jet printing (AJP) has emerged as a promising noncontact additive manufacturing method for high-resolution printing for a wide range of material systems. A key challenge limiting the broader adoption of AJP in the material science community is the lack of methods to precisely control thickness. Herein, we develop a model-based design of experiment (MBDoE) framework that integrates physics-informed models, nonlinear regression, and information criteria to postulate, select and calibrate the best model to describe and optimize the AJP manufacturing process. Starting with already available data from system commissioning (e.g., prior single variable sensitivity analysis), four candidate physics-informed models are postulated and trained. MBDoE identifies a single additional optimal experiment to validate these predictive models with quantified uncertainties, which are then used to determine the best experimental conditions to control printed film thickness. As a comparative benchmark, the analysis is repeated using the same dataset with nonparametric Gaussian process regression (GPR) model that does not incorporate physical information. Using MBDoE principles, we find that only five experiments are necessary to calibrate the nonlinear physics-informed parametric model, and with said limited data, this model outperforms the black-box machine learning GPR model. This key result underscores an emerging trend in the data science community: incorporating physical information into predictive models often drastically reduces the data requirements. Leveraging MBDoE further increased the data efficiency. By design, the proposed data science framework is general in nature and can be easily extended to other experimental and additive manufacturing systems beyond AJP.
Electroactive polymers (EAP) are an emerging class of actuation materials. Their large electrically induced strains (longitudinal or bending), low density, mechanical flexibility, and ease of processing offer advantages over traditional electroactive materials. However, before the capability of these materials can be exploited, their electrical and mechanical behavior must be properly quantified. Two general types of EAP can be identified. The first type is ionic EAP, which requires relatively low voltages (<10V) to achieve large bending deflections. This class usually needs to be hydrated and electrochemical reactions may occur. The second type is Electronic-EAP and it involves electrostrictive and/or Maxwell stresses. This type of materials requires large electric fields (>100MV/m) to achieve longitudinal deformations at the range from 4 - 360%. Some of the difficulties in characterizing EAP include: nonlinear properties, large compliance (large mismatch with metal electrodes), nonhomogeneity resulting from processing, etc. To support the need for reliable data, the authors are developing characterization techniques to quantify the electroactive responses and material properties of EAP materials. The emphasis of the current study is on addressing electromechanical issues related to the ion-exchange type EAP also known as IPMC. The analysis, experiments and test results are discussed in this paper.
A multiphase selfadaptive predictor corrector type algorithm was developed. This algorithm enables the solution of highly nonlinear structural responses including kinematic, kinetic and material effects as well as pro/post buckling behavior. The strategy involves three main phases: (1) the use of a warpable hyperelliptic constraint surface which serves to upperbound dependent iterate excursions during successive incremental Newton Ramphson (INR) type iterations; (20 uses an energy constraint to scale the generation of successive iterates so as to maintain the appropriate form of local convergence behavior; (3) the use of quality of convergence checks which enable various self adaptive modifications of the algorithmic structure when necessary. The restructuring is achieved by tightening various conditioning parameters as well as switch to different algorithmic levels to improve the convergence process. The capabilities of the procedure to handle various types of static nonlinear structural behavior are illustrated.
Fusion Transfer Learning (FTL) model provides a new paradigm to study high-dimensional dynamical behaviors, such as those in fusion plasma systems. The knowledge transfer process leverages a pre-trained neural encoder-decoder network, initially trained on linear simulations, to effectively capture nonlinear dynamics. The low-dimensional embeddings extract the coherent structures of interest, while preserving the inherent dynamics of the complex system. Experimental results highlight FTL's capacity to capture transitional behaviors and dynamical features in plasma dynamics -- a task often challenging for conventional methods. The model developed in this study is generalizable and can be extended broadly through transfer learning to address various magnetohydrodynamics (MHD) modes.
Physics-informed neural networks (PINNs) based machine learning is an emerging framework for solving nonlinear differential equations. However, due to the implicit regularity of neural network structure, PINNs can only find the flattest solution in most cases by minimizing the loss functions. In this paper, we combine PINNs with the homotopy continuation method, a classical numerical method to compute isolated roots of polynomial systems, and propose a new deep learning framework, named homotopy physics-informed neural networks (HomPINNs), for solving multiple solutions of nonlinear elliptic differential equations. The implementation of an HomPINN is a homotopy process that is composed of the training of a fully connected neural network, named the starting neural network, and training processes of several PINNs with different tracking parameters. The starting neural network is to approximate a starting function constructed by the trivial solutions, while other PINNs are to minimize the loss functions defined by boundary condition and homotopy functions, varying with different tracking parameters. These training processes are regraded as different steps of a homotopy process, and a PINN is initialized by the well-trained neural network of the previous step, while the first starting neural network is initialized using the default initialization method. Finally, several numerical examples are presented to show the efficiency of our proposed HomPINNs, including reaction-diffusion equations with a heart-shaped domain.
Most gas turbine combustion codes rely on ad-hoc statistical assumptions regarding the outcome of fuel atomization processes. The modeling effort proposed in this project is aimed at developing a realistic model to produce accurate predictions of fuel atomization parameters. The model involves application of the nonlinear stability theory to analyze the instability and subsequent disintegration of the liquid fuel sheet that is produced by fuel injection nozzles in gas turbine combustors. The fuel sheet is atomized into a multiplicity of small drops of large surface area to volume ratio to enhance the evaporation rate and combustion performance. The proposed model will effect predictions of fuel sheet atomization parameters such as drop size, velocity, and orientation as well as sheet penetration depth, breakup time and thickness. These parameters are essential for combustion simulation codes to perform a controlled and optimized design of gas turbine fuel injectors. Optimizing fuel injection processes is crucial to improving combustion efficiency and hence reducing fuel consumption and pollutants emissions.
The thermal-vacuum chamber calibration data from the Advanced Microwave Sounding Unit-A (AMSU-A) for NOAA-K, which will be launched in 1996, were analyzed to evaluate the instrument performance, including calibration accuracy, nonlinearity, and temperature sensitivity. The AMSU-A on NOAA-K consists of AMSU-A2 Protoflight Model and AMSU-A1 Flight Model 1. The results show that both models meet the instrument specifications, except the AMSU-A1 antenna beamwidths, which exceed the requirement of 3.3 +/- 10%. We also studied the instrument's radiometric characterizations which will be incorporated into the operational calibration algorithm for processing the in-orbit AMSU-A data from space. Particularly, the nonlinearity parameters which will be used for correcting the nonlinear contributions from an imperfect square-law detector were determined from this data analysis. It was found that the calibration accuracies (differences between the measured scene radiances and those calculated from a linear two-point calibration formula) are polarization-dependent. Channels with vertical polarizations show little cold biases at the lowest scene target temperature 84K, while those with horizontal polarizations all have appreciable cold biases, which can be up to 0.6K. It is unknown where these polarization-dependent cold biases originate, but it is suspected that some chamber contamination of hot radiances leaked into the cold scene target area. Further investigation in this matter is required. The existence and magnitude of nonlinearity in each channel were established and a quadratic formula for modeling these nonlinear contributions was developed. The model was characterized by a single parameter u, values of which were obtained for each channel via least-squares fit to the data. Using the best-fit u values, we performed a series of simulations of the quadratic corrections which would be expected from the space data after the launch of AMSU-A on NOAA-K. In these simulations, the cosmic background radiance corresponding to a cold space temperature 2.73K was adopted as one of the two reference points of calibration. The largest simulated nonlinear correction is about 0.3K, which occurs at channel 15 when the instrument temperature is at 38.09 deg C. Others are less than 0.2K in the remaining channels. Possible improvement for future instrument calibration is also discussed.
Under linear conditions, power injected from a single waveguide into a multi-core fiber array results in multimode propagation, progressively diminishing the spatial coherence of light. In this work, we introduce a comprehensive approach to mitigate this coherence loss by means of a nonlinear thermodynamic Joule–Thomson expansion. By leveraging the tools of optical thermodynamics, we demonstrate that as light undergoes a sudden transition from a small to a larger nonlinear optical array, it can abruptly drop its optical temperature to near-zero values. During this cooling process, light irreversibly flows into the system's fundamental mode with very high efficiency, synchronizing all elements of the lattice with the input port. We show that this nonlinear effect is highly predictable even in systems of arbitrary geometry and shape and can be controlled precisely by the initial conditions at the input of the array. In particular, for a single injection point, the reduction in optical temperature can be directly determined by the total power, irrespective of the input location.
Abstract Passive wave transformation via nonlinearity is ubiquitous in settings from acoustics to optics and electromagnetics. It is well known that different nonlinearities yield different effects on propagating signals, which raises the question of “what precise nonlinearity is the best for a given wave tailoring application?” In this work, considering a one-dimensional spring-mass chain connected by polynomial springs (a variant of the Fermi-Pasta-Ulam-Tsingou system), we introduce a bilevel inverse design method which couples the shape optimization of structures for tailored constitutive responses with reduced-order nonlinear dynamical inverse design. We apply it to two qualitatively distinct problems—minimization of peak transmitted kinetic energy from impact, and pulse shape transformation—demonstrating our method’s breadth of applicability. For the impact problem, we obtain two fundamental insights. First, small differences in nonlinearity can drastically change the dynamic response of the system, from severely under- to outperforming a comparative linear system. Second, the oft-used strategy of impact mitigation via “energy locking” bistability can be significantly outperformed by our optimal nonlinearity. We validate this case with impact experiments and find excellent agreement. This study establishes a framework for broader passive nonlinear mechanical wave tailoring material design, with applications to computing, signal processing, shock mitigation, and autonomous materials.