Reflections from artificial boundaries inhibit convergence of transient solutions to their steady limit. Far-field damping operators to suppress such reflections are presented for general first-order hyperbolic system...
详细信息
Reflections from artificial boundaries inhibit convergence of transient solutions to their steady limit. Far-field damping operators to suppress such reflections are presented for general first-order hyperbolic systems, and particular reference is made to the compressible Euler equations. The damping operator has the following properties: 1) no reflections are generated due to the introduction of the damping terms and 2) different wave systems may be damped at different rates. Feature 1 is imperative for successful damping and enables the attenuation of waves over relatively short length scales. Feature 2 enables the damping operator to act selectively on the outgoing waves alone, leaving the incoming waves unharmed. This property is desirable in genuine time-dependent problems where consistent information should be allowed to propagate from the artificial boundaries into the domain. Results for compressible Euler flows past airfoils show the potential of far-field damping in substantially accelerating convergence, particularly in fully subsonic problems.
The purpose of this paper is to consider the implementation of the parallel shooting for solving IVPs using a multiprocessor computing system. The paper shows the significance of these methods and demonstrates their e...
详细信息
The purpose of this paper is to consider the implementation of the parallel shooting for solving IVPs using a multiprocessor computing system. The paper shows the significance of these methods and demonstrates their effectiveness in controlling the numerical stability of the initial value algorithms.
Presented in this paper is an alternative design methodology for optimal discrete-time control-law synthesis (i..e, what we call a W-design method). The proposed method solves an optimal static output-feedback design ...
详细信息
Presented in this paper is an alternative design methodology for optimal discrete-time control-law synthesis (i..e, what we call a W-design method). The proposed method solves an optimal static output-feedback design for sampled-data systems with a given known sampling time. The overall scheme involves the formulation of an exact optimal control problem for discrete-time systems using a state-space transformation and an equivalent quadratic cost function in the w domain. A one-to-one relationship between a set of stabilizing feedback gains in the w domain and those in the traditional z domain has been established. The optimal output-feedback gain obtained from the minimization of the equivalent cost functional expressed in the w domain will yield the same optimum solution as the one obtained traditionally in the z domain. An attractive feature of this procedure is that design solutions can be obtained using only continuous-time-domain techniques. The procedure is demonstrated with the synthesis of a longitudinal-axis stability augmentation system for a commercial transport.
In this paper we derive a direct method for block tridiagonalizing a single-input single-output system triple {A, b, c}. The method is connected to the nonsymmetric Lanczos procedure developed in [10, 2, 1] and also l...
详细信息
In this paper we derive a direct method for block tridiagonalizing a single-input single-output system triple {A, b, c}. The method is connected to the nonsymmetric Lanczos procedure developed in [10, 2, 1] and also leads to canonical representations of such triples.
A locally implicit hybrid finite volume algorithm on mixed type of triangular and quadrilateral meshes for the two-dimensional steady and unsteady viscous flows has been developed. The unsteady, full Navier-Stokes equ...
详细信息
A locally implicit hybrid finite volume algorithm on mixed type of triangular and quadrilateral meshes for the two-dimensional steady and unsteady viscous flows has been developed. The unsteady, full Navier-Stokes equations are solved in the Cartesian coordinate system. A new construction of a total-variation-diminishing formulation on triangles is used to obtain high-resolution results in the convective-dominated flow region. In the viscous-dominated part, a second- and fourth-order dissipative model on quadrilateral grids is employed to minimize the numerical dissipation. To investigate the accuracy and characteristics of the present total-variation-diminishing scheme, the nonlinear Burgers' equation and inviscid flows passing through channels with sinusoidal bump in steady subsonic and unsteady transonic regimes are solved. The evaluation of the locally implicit hybrid algorithm for steady viscous flow is performed by studying the shock/boundary-layer interaction and transonic flows around a NACA 0012 airfoil and turbine cascade. To confirm the reliability and accuracy of the present numerical algorithm using time-accurate calculations, the viscous flow for an oscillating flat plate is investigated. Satisfactory agreement between numerical results and exact solutions for the Stokes' second problem is obtained.
This paper presents fixed-order dynamic compensator designs for the coupled mass benchmark problem. The design technique uses a method for loop transfer recovery at the plant input to approximately recover the transfe...
详细信息
This paper presents fixed-order dynamic compensator designs for the coupled mass benchmark problem. The design technique uses a method for loop transfer recovery at the plant input to approximately recover the transfer properties of an H(infinity) controller with full state feedback. The recovery is accomplished by the unique selection of quadratic performance index weighting matrices in a standard optimal output feedback formulation for fixed-order compensation. For the benchmark control problem, two compensator designs are developed. These designs use an observer canonical form to represent the compensator dynamics. The first design addresses problem 1, where only the spring stiffness is uncertain. The second design, a solution to problem 3, provides asymptotic rejection of a sinusoidal disturbance in the presence of an uncertain spring stiffness.
A corotational total Lagrangian formulation of beam element is presented for the nonlinear analysis of three-dimensional beam structures with large rotations but small strains. The nonlinear coupling among the bending...
详细信息
A corotational total Lagrangian formulation of beam element is presented for the nonlinear analysis of three-dimensional beam structures with large rotations but small strains. The nonlinear coupling among the bending, twisting, and stretching deformations is considered. All of the element deformations and equations are defined in body-attached element coordinates. Three rotation parameters are proposed to determine the orientation of the element cross section. Two sets of element nodal parameters termed explicit nodal parameters and implicit nodal parameters are introduced. The explicit nodal parameters are used in the assembly of the system equations from the element equations and chosen to be three components of the incremental translation vector and three components of the incremental rotation vector. The implicit nodal parameters are used to determine the deformations of the beam element and chosen to be three components of the total displacement vector and nodal values of the three rotation parameters. The element internal nodal forces corresponding to the implicit nodal parameters are obtained from the virtual work principle. numerical examples are presented and compared with the numerical and experimental results reported in the literature to demonstrate the accuracy and efficiency of the proposed method.
Angle-dependent spectral distortion (ASD) refers to the change in complex spectrum of the signal from the receiving transducer due to change in angular orientation of the reflecting target, i.e., from normal incidence...
详细信息
Angle-dependent spectral distortion (ASD) refers to the change in complex spectrum of the signal from the receiving transducer due to change in angular orientation of the reflecting target, i.e., from normal incidence to some non-normal incidence. The ASD for a fluid-fluid infinite planar interface (IP-ASD) is numerically determined for the cases of a planar piston, spherically focused piston, and nondiffracting (J0-Bessel) transducer geometries. The numerical algorithm employed is based on an equivalent image transducer. Reception of the acoustic pulse is performed using discrete surface integration, based on 3-D surface mesh generation strategies. numerical simulations, together with experimental results, are presented for the above transducers for various geometries, reflector distances, and misalignment angles. The results indicate that the IP-ASD significantly alters the spectrum of the received acoustic pulse, and thus must be taken into consideration for quantitative ultrasonic measurements. Finally, the numerical simulations have revealed that the nondiffracting transducer produces distinct peaks in the magnitude IP-ASD that enable precise estimates of misalignment angle to be obtained.
Some upwind schemes on stretched meshes are introduced. Our concept for numerical schemes is based on two relations: 1) a polynomial and the derivatives, and 2) a polynomial and the integrations. In this paper, the Ta...
详细信息
Some upwind schemes on stretched meshes are introduced. Our concept for numerical schemes is based on two relations: 1) a polynomial and the derivatives, and 2) a polynomial and the integrations. In this paper, the Taylor series expansion of a function q about a certain point and the Lagrangian interpolation formula are utilized to construct the upwind scheme on a physical space. Three results in two dimensions are shown: a motion of viscous vortex with a uniform background flow, a driven cavity flow, and a flow past a circular cylinder. The method presented produces smooth and realistic results for computations with stretched meshes.
We present a numerical method for the study of the stability-capacity diagram of the Ising perceptron that is readily scalable, thus providing the opportunity to resolve some open issues on this problem. We also discu...
详细信息
We present a numerical method for the study of the stability-capacity diagram of the Ising perceptron that is readily scalable, thus providing the opportunity to resolve some open issues on this problem. We also discuss the possibility of studying multi-layer systems.
暂无评论