Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Matrix equations”

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 1,279 records · Page 71
All sourcesNASA NTRSDOE OSTI
NASA NTRS1991

Efficient control with an order (n) recursive inversion of the Jacobian for an n-link serial manipulator

A method is presented to order (n) recursively invert the Jacobian matrix for a serial n-link manipulator. By realizing that the Jacobian relationship involving accelerations is a special case for the forward dynamics problem, it is possible to obtain the Jacobian inverse by directly applying Rodriguez's (1987) recursive forward dynamics algorithm. In this technique, the n-link robot equation is formulated as a spatially recursive algorithm in the form of a filtering and smoothing problem. To compute the Jacobian inverse via this algorithm, the n-link manipulator is modeled with only a unit end-point mass and zero applied torques.

Meldrum, D. R.↗
NASA NTRS1994

Analytic Sensitivities for Shape Optimization in Equivalent Plate Structural Wing Models

Equivalent plate modeling techniques based on Ritz analysis with simple polynomials prove to be efficient tools for structural modeling of wings in the preliminary design stage. Accuracy problems are encountered, however, when these models are used to obtain finite difference behavior sensitivities with respect to planform shape. The accuracy problems are associated with the poor numerical conditioning of static and eigenvalue equations. As higher-order polynomials are being used to Improve the analysis itself, the more sensitive is the finite difference derivative to the step size used. This article describes a formulation of wing equivalent plate modeling in which it is simple to obtain analytic, explicit expressions for stiffness and mass matrix elements without the need to perform numerical integration. This formulation leads naturally to analytic expressions for the derivatives of displacements, stresses, and natural frequencies with respect to shape design variables. This article examines the accuracy of finite difference derivatives compared with the analytic derivatives, and shows that In some cases it is impossible to obtain any information of value by finite differences. Analytic sensitivities, in this case, are still sufficiently accurate for design optimization.

Livne, Eli↗
NASA NTRS2012

Assessment of Polarization Effect on Efficiency of Levenberg-Marquardt Algorithm in Case of Thin Atmosphere Over Black Surface

The Levenberg-Marquardt algorithm [1, 2] provides a numerical iterative solution to the problem of minimization of a function over a space of its parameters. In our work, the Levenberg-Marquardt algorithm retrieves optical parameters of a thin (single scattering) plane parallel atmosphere irradiated by collimated infinitely wide monochromatic beam of light. Black ground surface is assumed. Computational accuracy, sensitivity to the initial guess and the presence of noise in the signal, and other properties of the algorithm are investigated in scalar (using intensity only) and vector (including polarization) modes. We consider an atmosphere that contains a mixture of coarse and fine fractions. Following [3], the fractions are simulated using Henyey-Greenstein model. Though not realistic, this assumption is very convenient for tests [4, p.354]. In our case it yields analytical evaluation of Jacobian matrix. Assuming the MISR geometry of observation [5] as an example, the average scattering cosines and the ratio of coarse and fine fractions, the atmosphere optical depth, and the single scattering albedo, are the five parameters to be determined numerically. In our implementation of the algorithm, the system of five linear equations is solved using the fast Cramer s rule [6]. A simple subroutine developed by the authors, makes the algorithm independent from external libraries. All Fortran 90/95 codes discussed in the presentation will be available immediately after the meeting from sergey.v.korkin@nasa.gov by request.

Korkin, S.↗
NASA NTRS2019

High Velocity Impact Performance of a Dual Layer Thermal Protection System for the Mars Sample Return Earth Entry Vehicle

The Mars Sample Return (MSR) Earth Entry Vehicle (EEV) is currently planned on being released from its Micro-Meteorite/Orbital Debri (MM/OD) shielding housing about two days before the Earth entry phase. This leaves the EEV exposed to incoming MM/OD impacts, potentially damaging the heat shield and compromising its Entry, Decent, and Landing (EDL) integrity. Currently, two materials are proposed to comprise the MSR-EEV heat shield, a dual layer material Heat-shield for Extreme Entry Environment Technology (HEEET) and Phenolic Impregnated Carbon Ablator (PICA). PICA has been well characterized for OD class impacts, ~7km/s high mass impacts, from previous testing done in the Orion program, but hasn't under-gone extensive MM impact testing. HEEET is a rela-tively new material with minimal prior testing in re-gards to High Velocity Impacts (HVI). In order to in-form selectability of a material, it is crucial to under-stand the material performance when faced with an HVI, directly affecting mission success probability.The current measure of a Thermal Protection Sys-tem's (TPS) performance against an HVI is evaluating a thermally sized material against its derived Balistic Limit Equation (BLE). A BLE is generated empirically from multiple shots of HVI testing, and is used as a first order method in evaluating TPS's performance against the expected MM/OD environment. This method has proved useful for previous uniform densi-ty TPS materials, but has never been validated against a dual layer recession material such as HEEET. Testing in the MSR program for FY19 has a re-quirement to assess the effects of a dual layer TPS by testing various thicknesses of HEEET's Recession Layer, seen in Figure 1. From this data, BLE's will be derived for the individual thickness ratio samples, as well as the material as a whole to evaluate if a heritage form of the BLE can capture the complex physics associated with a dual layer system. Crater morpholo-gy will also be assessed with post-test Non-Destructive Evaluation (NDE) methods such as CT scanning to visualize if a BLE can well predict the associated pene-tration depths, since a BLE assumes full disinigration of the impacting particle and is generally only used to size a spherical crater ? disregarding any shrapenel effects from a high density impactor.To inform this analysis, expected MM environ-ments from the Meteoroid Engineering Model (MEM) and analytical equations for mass flux of incoming MM were evaluated against the notional MSR-EEV trajectory [1]. Using those dispersions, a monte-carlo was run to determine the most probable particle pa-rameters, as well as the riskiest in terms of full bondline penetration. From these probabilities, a test matrix was designed to test against bounding cases for the various parameters of the BLE: projectile density, projectile mass, projectile velocity, and the impact angle.This presentation will discuss the performance of the dual layer TPS material HEEET against a wide range of impact kinetic energies and densitites, as well as the comparison of HEEET to PICA in terms of MM/OD performance and selectability criteria.

Libben, B. J.↗
NASA NTRS1971

A method for attitude control of a satellite in circular orbit by actively varying its inertia.

Equations are developed for a vehicle consisting of a system of particles which move in a generally prescribed fashion. These equations are then specialized to apply to a rigid vehicle with a constant amount of fluid contained in tanks and flowing through interconnecting lines. Various simplifications and specialized forms of the equations are derived, and relationships between mass and inertia that result from the fluid transfer logic and the physical constraints of the configuration are formulated. A tank arrangement and fluid transfer logic are devised which permit each vehicle product of inertia to be varied without varying the remaining elements of the inertia matrix or shifting the center of mass. Vehicle control is achieved by sensing angular rate and the local vertical and by transferring fluid as a function of these variables.

Clarke, W. D.↗
NASA NTRS1976

An investigation of flap-lag stability of wind turbine rotors in the presence of velocity gradients and helicopter rotors in forward flight

The flap-lag equations of motion of a torsionally rigid and centrally hinged spring-restrained rigid blade are developed using quasi-steady blade-element aerodynamic theory. These equations have periodic coefficients and are applicable to both wind-turbine rotors with velocity gradients and helicopter rotors in forward flight. By solving these equations both by the Floquet-Liapunov method and by an approximate method, the range of applicability of the latter method is established. Flap-lag and pure flapping stability boundaries illustrating the effects of flow and rotor parameters are presented. Finally, some discussion on the techniques for generating the Floquet transition matrix and on the relative merits of the coordinate axes system is presented.

Kaza, K. R. V.↗
NASA NTRS1978

An efficient finite element technique for sound propagation in axisymmetric hard wall ducts carrying high subsonic Mach number flows

The general non-linear three-dimensional equation for acoustic potential is derived by using a perturbation technique. The linearized axisymmetric equation is then solved by using a finite element algorithm based on the Galerkin formulation for a harmonic time dependence. The solution is carried out in complex number notation for the acoustic velocity potential. Linear, isoparametric, quadrilateral elements with non-uniform distribution across the duct section are implemented. The resultant global matrix is stored in banded form and solved by using a modified Gauss elimination technique. Sound pressure levels and acoustic velocities are calculated from post element solutions. Different duct geometries are analyzed and compared with experimental results.

Tag, I. A.↗
NASA NTRS1973

Energy partitioning of gaseous ions in an electric field.

The partitioning of ion energy among thermal energy, drift energy, and random-field energy is studied by solution of the Boltzmann equation. An expansion in powers of the square of the electric field strength is obtained by Kihara's method. Numerical calculations for several ion-neutral force laws show that Wannier's constant mean-free-time model gives a reasonable first approximation. The formal extension to multicomponent mixtures is also given. The matrix elements obtained are tabulated, and can be used to study the field dependence of other moments of the ion-distribution function.

Hahn, H.-S.↗
NASA NTRS1983

Research on the control of large space structures

The research effort on the control of large space structures at the University of Houston has concentrated on the mathematical theory of finite-element models; identification of the mass, damping, and stiffness matrix; assignment of damping to structures; and decoupling of structure dynamics. The objective of the work has been and will continue to be the development of efficient numerical algorithms for analysis, control, and identification of large space structures. The major consideration in the development of the algorithms has been the large number of equations that must be handled by the algorithm as well as sensitivity of the algorithms to numerical errors.

Denman, E. D.↗
NASA NTRS1992

CSM solutions of rotating blade dynamics using integrating matrices

The dynamic behavior of flexible rotating beams continues to receive considerable research attention as it constitutes a fundamental problem in applied mechanics. Further, beams comprise parts of many rotating structures of engineering significance. A topic of particular interest at the present time involves the development of techniques for obtaining the behavior in both space and time of a rotor acted upon by a simple airload loading. Most current work on problems of this type use solution techniques based on normal modes. It is certainly true that normal modes cannot be disregarded, as knowledge of natural blade frequencies is always important. However, the present work has considered a computational structural mechanics (CSM) approach to rotor blade dynamics problems in which the physical properties of the rotor blade provide input for a direct numerical solution of the relevant boundary-and-initial-value problem. Analysis of the dynamics of a given rotor system may require solution of the governing equations over a long time interval corresponding to many revolutions of the loaded flexible blade. For this reason, most of the common techniques in computational mechanics, which treat the space-time behavior concurrently, cannot be applied to the rotor dynamics problem without a large expenditure of computational resources. By contrast, the integrating matrix technique of computational mechanics has the ability to consistently incorporate boundary conditions and 'remove' dependence on a space variable. For problems involving both space and time, this feature of the integrating matrix approach thus can generate a 'splitting' which forms the basis of an efficient CSM method for numerical solution of rotor dynamics problems.

Lakin, William D.↗
NASA NTRS1972

Elimination on sparse symmetric systems of a special structure.

Consideration of the problem of finding a permutation of rows and columns and an algorithm for solving ordered systems of linear algebraic equations with sparse matrices having a certain regular structure. Two approaches to the solution of this problem, in which the sparsity is used to some extent, are outlined. One of them is a very general approach where optimal (or nearly optimal) ordering is sought and the algorithm for solving the ordered system treats the matrix element by element to perform only necessary operations. The other approach involves the use of band matrices. After comparing these two approaches, a third approach is then suggested which involves the use of pipe matrices, and a means of ordering the rows and columns to obtain this type of matrix is presented. Examples of matrices reordered by the proposed procedure are cited.

Segethova, J.↗
NASA NTRS1973

Decomposition-aggregation stability analysis

This report presents the development and description of the decomposition aggregation approach to stability investigations of high dimension mathematical models of dynamic systems. The high dimension vector differential equation describing a large dynamic system is decomposed into a number of lower dimension vector differential equations which represent interconnected subsystems. Then a method is described by which the stability properties of each subsystem are aggregated into a single vector Liapunov function, representing the aggregate system model, consisting of subsystem Liapunov functions as components. A linear vector differential inequality is then formed in terms of the vector Liapunov function. The matrix of the model, which reflects the stability properties of the subsystems and the nature of their interconnections, is analyzed to conclude over-all system stability characteristics. The technique is applied in detail to investigate the stability characteristics of a dynamic model of a hypothetical spinning Skylab.

Siljak, D. D.↗
NASA NTRS1980

Time-dependent difference theory for noise propagation in a two-dimensional duct

A time-dependent numerical formulation is derived for sound propagation in a two-dimensional straight soft-walled duct in the absence of mean flow. The time-dependent governing acoustic-difference equations and boundary conditions are developed along with the maximum stable time increment. Example calculations are presented for sound attenuation in hard- and soft-wall ducts. The time-dependent analysis has been found to be superior to the conventional steady numerical analysis because of much shorter solution times and the elimination of matrix storage requirements.

Baumeister, K. J.↗
NASA NTRS1991

Parallel Methods on Large-Scale Structural Analysis and Physics Applications; Symposium, Hampton, VA, Feb. 5, 6, 1991, Selected Papers

Recent advances in parallel methods and algorithms integrated into large-scale codes are presented. Consideration is given to problem decomposition (substructuring), efficient matrix solution algorithms for shared memory architectures, dynamic and transient analysis algorithms for shared memory architectures, and algorithms for distributed and massively parallel architectures. Particular attention is given to partitioning of unstructured problems for parallel processing, parallel-vector computation for linear-structural analysis and nonlinear unconstraint optimization problems, a parallel-vector equation solver for unsymmetric matrices on supercomputers, parallel nonlinear finite element dynamic response, multigrid algorithms for solving structural mechanics problems on supercomputers, structural analysis on massively parallel computers, explicit finite element methods with contact-impact on SIMD computers, and the impact of mapping and sparsity on parallelized finite element method modules.

Storaasli, Olaf O.↗
NASA NTRS2010

Development of a Response Surface Thermal Model for Orion Mated to the International Space Station

A study was performed to determine if a Design of Experiments (DOE)/Response Surface Methodology could be applied to on-orbit thermal analysis and produce a set of Response Surface Equations (RSE) that accurately predict vehicle temperatures. The study used an integrated thermal model of the International Space Station and the Orion Outer mold line model. Five separate factors were identified for study: yaw, pitch, roll, beta angle, and the environmental parameters. Twenty external Orion temperatures were selected as the responses. A DOE case matrix of 110 runs was developed. The data from these cases were analyzed to produce an RSE for each of the temperature responses. The initial agreement between the engineering data and the RSE predictions was encouraging, although many RSEs had large uncertainties on their predictions. Fourteen verification cases were developed to test the predictive powers of the RSEs. The verification showed mixed results with some RSE predicting temperatures matching the engineering data within the uncertainty bands, while others had very large errors. While this study to not irrefutably prove that the DOE/RSM approach can be applied to on-orbit thermal analysis, it does demonstrate that technique has the potential to predict temperatures. Additional work is needed to better identify the cases needed to produce the RSEs

Miller, Stephen W.↗
NASA NTRS2010

Micromechanics Modeling of Composites Subjected to Multiaxial Progressive Damage in the Constituents

The high-fidelity generalized method of cells composite micromechanics model is extended to include constituent-scale progressive damage via a proposed damage model. The damage model assumes that all material nonlinearity is due to damage in the form of reduced stiffness, and it uses six scalar damage variables (three for tension and three for compression) to track the damage. Damage strains are introduced that account for interaction among the strain components and that also allow the development of the damage evolution equations based on the constituent material uniaxial stress strain response. Local final-failure criteria are also proposed based on mode-specific strain energy release rates and total dissipated strain energy. The coupled micromechanics-damage model described herein is applied to a unidirectional E-glass/epoxy composite and a proprietary polymer matrix composite. Results illustrate the capability of the coupled model to capture the vastly different character of the monolithic (neat) resin matrix and the composite in response to far-field tension, compression, and shear loading.

Bednarcyk, Brett A.↗
NASA NTRS1974

Elementary solutions of coupled model equations in the kinetic theory of gases

The method of elementary solutions is employed to solve two coupled integrodifferential equations sufficient for determining temperature-density effects in a linearized BGK model in the kinetic theory of gases. Full-range completeness and orthogonality theorems are proved for the developed normal modes and the infinite-medium Green's function is constructed as an illustration of the full-range formalism. The appropriate homogeneous matrix Riemann problem is discussed, and half-range completeness and orthogonality theorems are proved for a certain subset of the normal modes. The required existence and uniqueness theorems relevant to the H matrix, basic to the half-range analysis, are proved, and an accurate and efficient computational method is discussed. The half-space temperature-slip problem is solved analytically, and a highly accurate value of the temperature-slip coefficient is reported.

Kriese, J. T.↗
NASA NTRS1985

Higher order methods for convection-diffusion problems

This paper applies C1 cubic Hermite polynomials embedded in an orthogonal collocation scheme to the spatial discretization of the unsteady nonlinear Burgers equation as a model of the equations of fluid mechanics. The temporal discretization is carried out by means of either a noniterative finite difference or an iterative finite difference procedure. Results of this method are compared with those of a second-order finite difference scheme and a splined-cubic Taylor's series scheme. Stability limits are derived and the matrix structure of the several schemes are compared.

Murphy, J. D.↗