Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/13628
Title: APPLICATION OF NEURAL NETWORKS FOR BLAST LOADING ON STRUCTURES
Authors: Venkateswarlu, N.
Keywords: EARTHQUAKE ENGINEERING
NEURAL NETWORKS
BLAST LOADING
ARTIFICIAL NEURAL NETWORKS
Issue Date: 2000
Abstract: In dealing with problems in analysis and design of civil engineering systems, we need to carryout rigorous and complex calculations. At present, with the help of computers, various emerging analytical tools are coming up to make the laborious process easy and at the same time conforming to a greater extent of reliability when compared to hand calculations. Artificial neural networks are one step ahead of conventional programming techniques, these networks simulates the working nature of human brain. The main advantage in the use of artificial neural networks is the capability of producing acceptable solutions even for situations with imprecise, imperfect and incomplete data also. In the present work, the application of artificial neural networks has been examined with specific reference to the Blast loading phenomenon on structures and developed four network models i.e., Pressure Net, Response Net, General Net and Height Net for evaluating the Design pressure and Response of the Structures. A comparative study has been carried out to see the variation of Design pressure and Response of structures depending on various parameters of Structure and its location from blast site. On the basis of studies conducted Neural Networks have been found to perform to a high degree of accuracy. For the more Neural Networks can be useful employed to conduct parametric studies to argument our knowledge of the blast phenomenon.
URI: http://hdl.handle.net/123456789/13628
Other Identifiers: M.Tech
Appears in Collections:MASTERS' DISSERTATIONS (Earthquake Engg)

Files in This Item:
File Description SizeFormat 
EQD 248438.pdf7.62 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.