On certain approximations in the finite- element method
Finite element method approximations outside variational principle in elasticity problems, examining convergence to exact solution
SEARCH · Search NASA
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.
Finite element method approximations outside variational principle in elasticity problems, examining convergence to exact solution
Convergent finite element equations for dynamic stability of plates dependent on vibration and buckling modes
Finite element analyses characterized by monotonic convergence include the discipline for meaningful measurements of convergence rate and consequently economical extrapolation. Few proposers of element models guarantee monotonic convergence for their elements. Thus, a need exists for an automatic test to classify available element models. This paper describes such a test - a test can be performed using a digital computer to guarantee that a particular element model imbues monotonicity. It describes the test and its basis. It examines seven element models for a rectangular membrane to illustrate the value of the tests. Besides confirming results already known, the application yields new data. It 'proves' monotonicity for two improved models, defines the range of element proportions for which another element can be guaranteed to exhibit monotonicity, and suggests that another element is deficient. In the special case of absolutely convergent membrane displacement models, proof of monotonicity is a necessary and sufficient condition to insure that upper bound estimates of strain energy are developed. Accordingly, the test furnishes a proof of bound solutions independently of requirements on displacement continuity the element basis may or may not satisfy.
Finite element equations are developed for studying surface temperatures resulting from frictional heating in sliding systems. The equations include the effect of velocity of moving components, an effect which is found to be quite significant, even at low sliding velocities. A program was written using the equations and it was applied to the study of surface temperatures in two different sliding systems: dry or boundary lubricated sleeve bearings and a labyrinth gas path seal configuration. Very good agreement was achieved between analytical predictions using the program and experimental temperature measurements. The program was used to study the influence of various material parameters on surface temperatures in the two sliding systems.
Finite element methods are developed for a one-dimensional singular perturbation problem involving a nonlinear flux function. In the limit of vanishing dissipation, the numerical flux is found to be almost identical to that due to Engquist and Osher, the only difference being in the treatment of transonic compression. This facilitates sharper resolution of stationary shocks, but is accompanied by loss of smoothness and monotonicity.
Structural finite elements have been used to model the coupled fluidic-structural response of a liquid oxygen (LOX) feedline at a rocket engine test facility. The model simulates the effects of Pogo pulsing, a test procedure which uses a piston in a side branch to impart an oscillatory pressure pulse to the LOX column as it feeds the engine. In addition to the feedline's structural characteristics, the model accounts for the mass and axial stiffness of the fluid column, the oscillatory pulse of the piston, and the hydraulic impedance of the rocket engine. The model was used to determine the relations between piston stroke, pressure oscillation at the engine inlet, and structural excitation of the feedline. This paper develops the concepts employed by the model.
The finite element model of a geared rotor system on flexible bearings has been developed. The model includes the rotary inertia of shaft elements, the axial loading on shafts, flexibility and damping of bearings, material damping of shafts and the stiffness and the damping of gear mesh. The coupling between the torsional and transverse vibrations of gears were considered in the model. A constant mesh stiffness was assumed. The analysis procedure can be used for forced vibration analysis of geared rotors by calculating the critical speeds and determining the response of any point on the shaft to mass unbalances, geometric eccentricities of gears and displacement transmission error excitation at the mesh point. The dynamic mesh forces due to these excitations can also be calculated. The model has been applied to several systems for the demonstration of its accuracy and for studying the effect of bearing compliances on system dynamics.
A finite element formulation and solution procedures is presented for limit-cycle motions of 2D panels subjected simultaneously to thermal and aerodynamic loads. The thermal load is described by a steady-state temperature distribution and the quasi-steady first-order piston theory is used for aerodynamic pressure. The von Karman nonlinear strain-displacement relationship is used for the large panel deflections. Three temperature distributions are evaluated: (1) uniform; (2) symmetric sinusoidal varying temperature along panel length; and (3) linearly varying temperature through panel thickness. The influence of these three temperature distributions on limit-cycle motions and flutter boundaries of a simply supported 2D panel is investigated.
Finite Element Models (FEM's) are used in the design and analysis of aircraft to mathematically describe the airframe structure for such diverse tasks as flutter analysis and actively controlled landing gear design. FEM's are used to model the entire airplane as well as airframe components. The purpose of this document is to describe recommended methods for verifying the quality of the FEM's and to specify a step-by-step procedure for implementing the methods.
A finite element model of a geared rotor system on flexible bearings has been developed. The model includes the rotary inertia of on shaft elements, the axial loading on shafts, flexibility and damping of bearings, material damping of shafts and the stiffness and the damping of gear mesh. The coupling between the torsional and transverse vibrations of gears were considered in the model. A constant mesh stiffness was assumed. The analysis procedure can be used for forced vibration analysis geared rotors by calculating the critical speeds and determining the response of any point on the shafts to mass unbalances, geometric eccentricities of gears, and displacement transmission error excitation at the mesh point. The dynamic mesh forces due to these excitations can also be calculated. The model has been applied to several systems for the demonstration of its accuracy and for studying the effect of bearing compliances on system dynamics.
A finite element velocity potential program was developed to study acoustic wave propagation in complex geometries. For irrotational flows, relatively low sound frequencies, and plane wave input, the finite element solutions showed significant effects of inlet curvature and flow gradients on the attenuation of a given acoustic liner in a realistic variable area turbofan inlet. The velocity potential approach can not be used to estimate the effects of rotational flow on acoustic propagation, since the potential acoustic disturbances propagate at the speed of the media in sheared flow. Approaches are discussed that are being considered for extending the finite element solution to include the far field, as well as the internal portion of the duct. A new matrix partitioning approach is presented that can be incorporated in previously developed programs to allow the finite element calculation to be marched into the far field. The partitioning approach provided a large reduction in computer storage and running times.
A finite element velocity potential program has been developed to study acoustic wave propagation in complex geometries. For irrotational flows, relatively low sound frequencies, and plane wave input, the finite element solutions show significant effects of inlet curvature and flow gradients on the attenuation of a given acoustic liner in a realistic variable area turbofan inlet. In addition, as shown in the paper, the velocity potential approach can not be used to estimate the effects of rotational flow on acoustic propagation since the potential acoustic disturbances propagate at the speed of the media in sheared flow. Approaches are discussed that are being considered for extending the finite element solution to include the far field as well as the internal portion of the duct. A new matrix partitioning approach is presented that can be incorporated in previously developed programs to allow the finite element calculation to be marched into the far field. The partitioning approach provides a large reduction in computer storage and running times.
Finite elements in time as an alternative strategy for rotorcraft trim problems are studied. The research treats linear flap and linearized flap-lag response both for quasi-trim and trim cases. The connection between Fourier series analysis and hp-finite elements for periodic a problem is also examined. It is proved that Fourier series is a special case of space-time finite elements in which one element is used with a strong displacement formulation. Comparisons are made with respect to accuracy among Fourier analysis, displacement methods, and mixed methods over a variety parameters. The hp trade-off is studied for the periodic trim problem to provide an optimum step size and order of polynomial for a given error criteria. It is found that finite elements in time can outperform Fourier analysis for periodic problems, and for some given error criteria. The mixed method provides better results than does the displacement method.
Finite elements in time as an alternative strategy for rotorcraft trim problems are studied. The research treats linear flap and linearized flap-lag response both for quasi-trim and trim cases. The connection between Fourier series analysis and hp-finite elements for periodic a problem is also examined. It is proved that Fourier series is a special case of space-time finite elements in which one element is used with a strong displacement formulation. Comparisons are made with respect to accuracy among Fourier analysis, displacement methods, and mixed methods over a variety parameters. The hp trade-off is studied for the periodic trim problem to provide an optimum step size and order of polynomial for a given error criteria. It is found that finite elements in time can outperform Fourier analysis for periodic problems, and for some given error criteria. The mixed method provides better results than does the displacement method.
The stress-intensity factors for most of the commonly used fracture specimens (center-crack tension, single and double edge-crack tension, and compact), those that have a through-the-thickness crack, were calculated using a three dimensional finite-element elastic stress analysis. Three-dimensional singularity elements were used around the crack front. The stress intensity factors along the crack front were evaluated by using a force method, developed herein, that requires no prior assumption of either plane stress or plane strain. The calculated stress-intensity factors from the present analysis were compared with those from the literature whenever possible and were generally found to be in good agreement. The stress-intensity factors at the midplane for all specimens analyzed were within 3 percent of the two dimensional plane strain values. The stress intensity factors at the specimen surfaces were considerably lower than at the midplanes. For the center-crack tension specimens with large thickness to crack-length ratios, the stress-intensity factor reached a maximum near the surface of the specimen. In all other specimens considered the maximum stress intensity occurred at the midplane.
In the Probabilistic Finite Element Method (PFEM), finite element methods have been efficiently combined with second-order perturbation techniques to provide an effective method for informing the designer of the range of response which is likely in a given problem. The designer must provide as input the statistical character of the input variables, such as yield strength, load magnitude, and Young's modulus, by specifying their mean values and their variances. The output then consists of the mean response and the variance in the response. Thus the designer is given a much broader picture of the predicted performance than with simply a single response curve. These methods are applicable to a wide class of problems, provided that the scale of randomness is not too large and the probabilistic density functions possess decaying tails. By incorporating the computational techniques we have developed in the past 3 years for efficiency, the probabilistic finite element methods are capable of handling large systems with many sources of uncertainties. Sample results for an elastic-plastic ten-bar structure and an elastic-plastic plane continuum with a circular hole subject to cyclic loadings with the yield stress on the random field are given.
A finite element method is presented for the large amplitude vibrations of complex structures that can be modelled with beam and rectangular plate elements subjected to harmonic excitation. Both inplane deformation and inertia are considered in the formulation. Derivation of the harmonic force and nonlinear stiffness matrices for a beam and a rectangular plate element are presented. Solution procedures and convergence characteristics of the finite element method are described. Nonlinear response to uniform and concentrated harmonic loadings and improved nonlinear free vibration results are presented for beams and rectangular plates of various boundary conditions.
Finite Element in Time has been proven to be a powerful alternative solving strategy for the rotor craft trim problem. Additionally, Finite Element Method in Time has been developed in various versions like time-marching framework, Galerkin framework, Rayleigh-Ritz framework, and mixed formulation. Recently, this method was applied to the rotorcraft trim problem to obtain linearized solutions. The rotorcraft trim problem consists of trying to find a period solution for period-coefficient, differential equations subject to side constraints where certain force and momentum balance equations are forced to be equal to zero. There are free (or trim) parameters that are chosen to meet these side constraints. This project aims at expanding the application, in terms of the rotorcraft trim problem, from a linearized solution to nonlinear solution.