Problems on the identification of two-dimensional spatial domains arising in the detection and characterization of structural flaws in materials are considered. For a thermal diffusion system with external boundary in...
详细信息
Problems on the identification of two-dimensional spatial domains arising in the detection and characterization of structural flaws in materials are considered. For a thermal diffusion system with external boundary input, observations of the temperature on the surface are used in an output least square approach. Parameter estimation techniques based on the method of mappings are discussed, and approximation schemes are developed based on a finite-element Galerkin approach. Theoretical convergence results for computational techniques are given, and the results are applied to the identification of two kinds of boundary shapes.< >
A compact difference scheme derived in (Phillips and Rose, SIAM J. Statist. Comput. 7 , 288 (1986)) for treating the equilibrium equations of elasticity is studied. The scheme truns out to be inconsistent and unstable...
A compact difference scheme derived in (Phillips and Rose, SIAM J. Statist. Comput. 7 , 288 (1986)) for treating the equilibrium equations of elasticity is studied. The scheme truns out to be inconsistent and unstable. A multigrid method which takes into account these properties is described. The solution of the discrete equations, up to the level of discretization errors, is obtained by this method in just 16 work units, where a work unit is the work involved in relaxing the finest grid equations once.
This report describes distributed implementations of the red/black SOR algorithm and of the MGR[ν] multigrid algorithm on the Crystal multicomputer. Rates of convergence and observed efficiencies for both algorithms ...
详细信息
This report describes distributed implementations of the red/black SOR algorithm and of the MGR[ν] multigrid algorithm on the Crystal multicomputer. Rates of convergence and observed efficiencies for both algorithms are compared.
Numerical experiments are performed to compare the accuracy obtained when physical and transform space filters are used to smooth the oscillations in Fourier collocation approximations to discontinuous solutions of a ...
详细信息
In this paper we propose a fast method for solving wave guide problems. In particular, we consider the guide to be inhomogeneous, and allow propagation of waves of higher-order modes. Such techniques have been handled...
详细信息
Neural networks have attracted much interest recently, and using parallel architectures to simulate neural networks is a natural and necessary application. The SIMD model of parallel computation is chosen, because sys...
Neural networks have attracted much interest recently, and using parallel architectures to simulate neural networks is a natural and necessary application. The SIMD model of parallel computation is chosen, because systems of this type can be built with large numbers of processing elements. However, such systems are not naturally suited to generalized communication. A method is proposed that allows an implementation of neural network connections on massively parallel SIMD architectures. The key to this system is an algorithm that allows the formation of arbitrary connections between the "neurons". A feature is the ability to add new connections quickly. It also has error recovery ability and is robust over a variety of network topologies. Simulations of the general connection system, and its implementation on the Connection Machine, indicate that the time and space requirements are proportional to the product of the average number of connections per neuron and the diameter of the interconnection network.
Due to demographical factors, there will be a 25% decline in the national labor pool of eligible 17 to 21 year old men by 1992. As the Navy faces fierce competition with other services and private industry for the dwi...
详细信息
Due to demographical factors, there will be a 25% decline in the national labor pool of eligible 17 to 21 year old men by 1992. As the Navy faces fierce competition with other services and private industry for the dwindling personnel resources of the nation, efforts must be made to consider automating certain tasks, especially those that are hazardous, boring, and labor intensive onboard ships at sea. The Surface Ship Continuing Concept Formulation (CONFORM) Program (SEA-5014) sponsored the Naval Surface Weapons Center's Robotics Laboratory to identify potential applications of robotic and artificial intelligence systems to operation and mission activities for shipboard use. The results of the investigation concerning applications of robotics to automate the aforementioned tasks are presented. Changes in current manning levels for selected tasks and ship classes, i.e., CVs, CVNs, LPDs, LPNs, and auxiliaries are examined. Future ship designs incorporating robotics, such as the advance based repair (ABR) ship are discussed. Twenty-four applications were cited in the study which include a remote controlled vehicle for flight deck operations, limited tasks in biological, chemical, and radiation environments, computer aided command decision aids, ordnance handling, undersea search, recovery, and salvage, and ventilation duct cleaning. Four of the applications and their subsequent hardware fruition will be discussed.
In this paper we study an inviscid model for a steady axisymmetric flow with siwrl. The governing equation is a nonlinear elliptic equation which has more than one solution for a certain range of the swirl parameter. ...
详细信息
In this paper we study an inviscid model for a steady axisymmetric flow with siwrl. The governing equation is a nonlinear elliptic equation which has more than one solution for a certain range of the swirl parameter. The physically interesting solutions have closed streamlines that look like vortex breakdown (‘bubble’-like solutions). A multigrid method is used to find these solutions. Using an FMG algorithm (nested iteration), the problem is solved in just a few multigrid cycles.
We describe a compact finite difference scheme to solve div v − q 2 u = 0, v = ϱ grad u , which does not depend upon first eliminating v , so as to obtain a single, higher-order equation. The finite difference equatio...
We describe a compact finite difference scheme to solve div v − q 2 u = 0, v = ϱ grad u , which does not depend upon first eliminating v , so as to obtain a single, higher-order equation. The finite difference equations can be efficiently solved by standard iterative methods, however, by first algebraically eliminating the flux components from the compact form. Energy estimates show that u is approximated to second-order accuracy.
A nonconforming finite element method is described for treating linear equilibrium problems, and a convergence proof showing second order accuracy is given. The close relationship to a related compact finite differenc...
暂无评论