دانلود رایگان مقاله لاتین سست سازی پردازش گرافیک از سایت الزویر
عنوان فارسی مقاله:
سست سازی واحد پردازش گرافیک برای نوار تیرها و عناصر
عنوان انگلیسی مقاله:
Vectorised graphics processing unit accelerated dynamic relaxation for bar and beam elements
سال انتشار : 2016
برای دانلود رایگان مقاله سست سازی پردازش گرافیک اینجا کلیک نمایید.
مقدمه انگلیسی مقاله:
1. Introduction
Dynamic relaxation (DR) [1,2] is a dynamic analysis procedure where one is interested in the final rest state of the structure, that is, the static solution after the dynamic oscillations have been suf- ficiently damped. The structural system is considered to be in equilibrium after a convergence criterion is satisfied, for example, when the balance of internal and external forces is suitably close relative to some tolerance. The initial geometry of the structure need not be close to the final rest state, and so the process may be initialised from an approximate geometry and found after the external loads are instantaneously applied. The method is particularly adept at solving large displacement problems and has found application withanalysing assemblies of cable, bending and membrane elements, and is used extensively for tensile structures [3]. There are various examples in the literature of using DR to analyse beam elements. Work related to bending elements stabilised by membranes can be found in [4]; in [5], six degree-of-freedom elements were used to analyse the erection and loading of double-layered timber beams for grid-shells, with an optimisation method for variable crosssection depths; tubular glass-fibre-reinforced grid-shells have been investigated by [6], and [7] looked at bending elements supported by cables and/or cable-nets. Although membrane elements are not investigated in this research, constant strain triangular elements could be used, as formulations in the literature describe the behaviour of the membrane element via the edges of the faces to give an equivalent cable-net system. Therefore by creating a relationship between membrane stresses and edge link tensile forces, the methods described here-in for net structures may be extended readily for application to membrane elements. Further literature on membrane elements in the dynamic relaxation method can be found in [2,8,9] and [10].
برای دانلود رایگان مقاله سست سازی پردازش گرافیک اینجا کلیک نمایید.
کلمات کلیدی:
General-purpose computing on graphics processing units - Wikipedia https://en.wikipedia.org/.../General-purpose_computing_on_graphics_processing_unit... General-purpose computing on graphics processing units is the use of a graphics processing .... MATLAB supports GPGPU acceleration using the Parallel Computing Toolbox and MATLAB ... Most operations on the GPU operate in a vectorized fashion: one operation can be performed on up to four values at once. GPU accelerated sparse matrix-vector multiplication and sparse matrix ... onlinelibrary.wiley.com/doi/10.1002/cpe.3415/full by Y Tao - 2015 - Cited by 6 - Related articles Oct 7, 2014 - Summary. Many high performance computing applications require computing both sparse matrix-vector product (SMVP) and sparse ... XIV Mediterranean Conference on Medical and Biological Engineering ... https://books.google.com/books?isbn=3319327038 Efthyvoulos Kyriacou, Stelios Christofides, Constantinos S. Pattichis - 2016 - Technology & Engineering In terms of bit-vector operations, the algorithm is accelerated by using the ... data (SIMD) units, (b) using the graphics processing unit (GPU) and (c) using a ... Machine Learning, Optimization, and Big Data: First International ... https://books.google.com/books?isbn=3319279262 Panos Pardalos, Mario Pavone, Giovanni Maria Farinella - 2016 - Computers Few GPU-based implementations have been proposed for support vector ... [24] make use of this implementation to accelerate a local search scheme that ... Electronic Structure Calculations on Graphics Processing Units: From ... https://books.google.com/books?isbn=1118661788 Ross C. Walker, Andreas W. Goetz - 2016 - Science As discussed in Chapter 2 the rise of GPUs was due to the relative mismatch between ... Over time, however, the increasing thread parallelism of SIMD vector units in CPUs ... for performance acceleration when porting these methods to GPUs.
