No abstract

Tall buildings are slender flexible structures and need to be investigated to ascertain the importance of wind induced oscillations or excitation along and across the direction of wind. The Indian code of practice for wind load on buildings and structures (IS-875 Part-3 1987) gives a procedure to determine along wind response of tall structures, while the across wind response and interference effect are not included in the code at present. A document 'Review of Indian Wind Code IS 875 (Part 3) 1987' has been prepared by IIT Kanpur under GSDMA project gives recommendations to obtain across wind response of tall buildings as per procedure given in Australian/New Zealand standard 'Structural Design Actions -Part 2 Wind Action (AS/NZS 1170-2 : 2002) In the Australian code to obtain the cross wind response it is necessary to compute the coefficient (Cfs) for which figures and expressions are given for selected (h:b:d) ratios. In this paper use of Artificial Neural Network (ANN) has been made to generalize the above procedure from the limited available data, so that across wind response can be obtained for a building with given (h:b:d) ratio.

No abstract

No abstract

Fixed-grid discretization strategy is proposed for static structural Finite Element Analysis (FEA) of Functionally Graded Materials (FGM). The fixed-grid strategy reduces numerical integration cost dramatically by generating a single local stiffness matrix for isotropic materials. For FGMs, domain is discretized into the layers in such a way that material properties in each layer are constant. Therefore, for each layer, a single local stiffness matrix will be constructed. These matrices are directly used in the solver phase of the assembly-free FEM without constructing the global stiffness matrix. The fixed-grid strategy reduces the global memory transactions on the GPU by storing these elemental matrices in on-chip shared memory or cached constant memory.Furthermore, the assembly-free method is adopted to leverage a fine grained parallelism on the GPU at the degree of freedom level. Numerical experiments showed the effectiveness of the discrete layered approach for FGM using the fixed-grid strategy. For performance evaluation two strategies using global memory and shared memory are compared and found that the use of shared memory can achieve approximately 2.4 times better performance than global memory.