skip to main content

Explore Research Products in the NSF-PAR

Title: Computational Modeling of Nonlinear Dynamics and Its Utility in MEMS Gyroscopes.
This paper describes a hybrid approach for modeling nonlinear vibrations and determining essential (normal form) coefficients that govern a reduced-order model of a structure. Incorporating both computational and analytical tools, this blended method is demonstrated by considering a micro-electro-mechanical vibrating gyroscopic rate sensor that is actuated by segmented DC electrodes. Two characterization methods are expatiated, where one is more favorable in computational tools and the other can be used in experiments. Using the reduced model, it is shown that tuning the nonuniform DC bias results in favorable changes in Duffing and mode-coupling nonlinearities which can improve the gyroscope angular rate sensitivity by two orders of magnitude.  more » « less
Award ID(s):
1662619
NSF-PAR ID:
10386062
Author(s) / Creator(s):
; ;
Date Published:
Journal Name:
Journal of Structural Dynamics
ISSN:
2684-6500
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ( , ASME Fluids Engineering Division Summer Meeting FEDSM2021)
    Hydrodynamic performance of ships can be greatly improved by the formation of air cavities under ship bottom with the purpose to decrease water friction on the hull surface. The air-cavity ships using this type of drag reduction are usually designed for and typically effective only in a relatively narrow range of speeds and hull attitudes and sufficient rates of air supply to the cavity. To investigate the behavior of a small-scale air-cavity boat operating under both favorable and detrimental loading and speed conditions, a remotely controlled model hull was equipped with a data acquisition system, video camera and onboard sensors to measure air-cavity characteristics, air supply rate and the boat speed, thrust and trim in operations on open-water reservoirs. These measurements were captured by a data logger and also wirelessly transmitted to a ground station and video monitor. The experimental air-cavity boat was tested in a range of speeds corresponding to length Froude numbers between 0.17 and 0.5 under three loading conditions, resulting in near zero trim and significant bow-up and bow-down trim angles at rest. Reduced cavity size and significantly increased drag occurred when operating at higher speeds, especially in the bow-up trim condition. The other objective of this study was to determine whether computational fluid dynamics simulations can adequately capture the recorded behavior of the boat and air cavity. A computational software Star-CCM+ was utilized with the VOF method employed for multi-phase flow, RANS approach for turbulence modeling, and economical mesh settings with refinements in the cavity region and near free surface. Upon conducting the mesh verification study, several experimental conditions were simulated, and approximate agreement with measured test data was found. Adaptive mesh refinement and time step controls were also applied to compare results with those obtained on the user-generated mesh. Adaptive controls improved resolution of complex shedding patterns from the air cavity but had little impact on overall results. The presented here experimental approach and obtained results indicate that both outdoor experimentation and computationally inexpensive modeling can be used in the process of developing air-cavity systems for ship hulls. 
    more » « less
  2. ; ; ; ; ; ( , 2021 37th Semiconductor Thermal Measurement, Modeling & Management Symposium (SEMI-THERM))
    null (Ed.)
    To reproduce a Digital Twin (DT) of a data center (DC), input data is required which is collected through site surveys. Data collection is an important step since accurate representation of a DC depends on capturing the necessary detail for various model fidelity levels of each DC component. However, guidance is lacking in this regard as to which components within the DC are crucial to achieve the level of accuracy desired for the computational model. And determining the input values of the component object parameters is an exercise in engineering judgement during site survey. Sensitivity analysis can be an effective methodology to determine how the level of simplification in component models can affect the model accuracy.In this study, a calibrated raised-floor DC model is used to study the sensitivity of a DC component's representation to the DC model accuracy. Commercial CFD tool, 6SigmaDC Room is used for modeling and simulation. A total of 8 DC components are considered and eventually ranked on the basis of time and effort required to collect model input data. For parametrized component object, the object's full range of input parameter values are considered, and simulations run. The results are compared with the baseline calibrated model to understand the trade-off between survey effort/cost and model accuracy. For the calibrated DC model and of the 8 components considered, it was observed that the chilled water piping branches, data cables and the cable penetration seal (found within cabinets) have considerable influence on the tile flow rate prediction accuracy. 
    more » « less
  3. ; ; ; ; ( , Microprocessors and Microsystems)
    Since many Cyber–Physical Systems (CPS) interact with the real world, they are safety- or mission- critical. Temporal specification languages like STL (Signal Temporal Logic) have been developed to capture the properties that built CPS must meet. However, the existing temporal logics/languages do not provide a natural way to express the tolerance with which the timing properties must be met. As a consequence of this, the specified properties may be vague, the ensuing CPS design may end up being over- or under-provisioned, and the validation of whether the built CPS meets the specified CPS properties may turn out to be erroneous. To address these issues, a run-time verification methodology is proposed, that allows users to explicitly specify the tolerance with which timing properties must be met. To ensure the correctness of measurement-based validation of a built CPS, this article: (i) proposes a test to determine if a given measurement system can validate the properties specified in TTL, and (ii) proposes a measurement-based testing methodology to provide one-sided guarantee that the built CPS meets the specified CPS properties. The guarantees are one-sided in the sense that when the measurement-based testing concludes that the properties are met, then they are guaranteed to be met (so not false positive). However, when the measurement-based testing concludes that the properties were not met, then they may have met (there can be false negative). In order to validate our claims, we built a model of flying paster (part of the printing press that swaps in a new roll of paper when the current roll is about to finish) using Arduino Mega 2560 and two Hansen brushed DC motors and specified the timing constraints among the various events in this system, along with the tolerances with which they should be met in TTL. We generated the testing logic and validated that we get no false positive, even though we encounter 4.04% false negative. The rate of false negatives can be reduced to be less than any arbitrary value by using more accurate measurement equipment. 
    more » « less
  4. ; ; ; ; ( , Inverse Problems)
    Abstract Obtaining lightweight and accurate approximations of discretized objective functional Hessians in inverse problems governed by partial differential equations (PDEs) is essential to make both deterministic and Bayesian statistical large-scale inverse problems computationally tractable. The cubic computational complexity of dense linear algebraic tasks, such as Cholesky factorization, that provide a means to sample Gaussian distributions and determine solutions of Newton linear systems is a computational bottleneck at large-scale. These tasks can be reduced to log-linear complexity by utilizing hierarchical off-diagonal low-rank (HODLR) matrix approximations. In this work, we show that a class of Hessians that arise from inverse problems governed by PDEs are well approximated by the HODLR matrix format. In particular, we study inverse problems governed by PDEs that model the instantaneous viscous flow of ice sheets. In these problems, we seek a spatially distributed basal sliding parameter field such that the flow predicted by the ice sheet model is consistent with ice sheet surface velocity observations. We demonstrate the use of HODLR Hessian approximation to efficiently sample the Laplace approximation of the posterior distribution with covariance further approximated by HODLR matrix compression. Computational studies are performed which illustrate ice sheet problem regimes for which the Gauss–Newton data-misfit Hessian is more efficiently approximated by the HODLR matrix format than the low-rank (LR) format. We then demonstrate that HODLR approximations can be favorable, when compared to global LR approximations, for large-scale problems by studying the data-misfit Hessian associated with inverse problems governed by the first-order Stokes flow model on the Humboldt glacier and Greenland ice sheet. 
    more » « less
  5. ; ; ; ; ( , Journal of Applied Physics)
    In plasma-driven solution electrolysis (PDSE), gas-phase plasma-produced species interact with an electrolytic solution to produce, for example, nanoparticles. An atmospheric pressure plasma jet (APPJ) directed onto a liquid solution containing a metallic salt will promote reduction of metallic ions in solution, generating metallic clusters that nucleate to form nanoparticles. In this article, results from a computational investigation are discussed of a PDSE process in which a radio-frequency APPJ sustained in helium impinges on a silver nitrate solution, resulting in growth of silver nanoparticles. A reaction mechanism was developed and implemented in a global plasma chemistry model to predict nanoparticle growth. To develop the reaction mechanism, density functional theory was used to generate probable silver growth pathways up to Ag 9 . Neutral clusters larger than Ag 9 were classified as nanoparticles. Kinetic reaction rate coefficients for thermodynamically favorable growth pathways were estimated based on an existing, empirically determined base reaction mechanism for smaller Ag particle interactions. These rates were used in conjunction with diffusion-controlled reaction rate coefficients that were calculated for other Ag species. The role of anions in reduction of Ag n ions in forming nanoparticles is also discussed. Oxygen containing impurities or admixtures to the helium, air entrainment into the APPJ, and dissociation of saturated water vapor above the solution can produce additional reactive oxygen species in solution, resulting in the production of anions and [Formula: see text] in particular. For a given molarity, delivering a sufficient fluence of reducing species will produce similar nanoparticle densities and sizes for all applied power levels. Comparisons are made to alternate models for nanoparticle formation, including charged nanoparticles and use of direct current plasmas. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email