The Structural Engineer > Archive > Volume 75 (1997) > Issues > Issue 14 > Correspondence on An Introduction to Neural Computing for the Structural Engineer by Professor W.M.
Name of File 6135-75-14.pdf cached at 15/12/2017 00:57:11 - with 1 pages. pdfPath: E:\k9.istructe.org\CMS\webtest\files\dd\dd6fb1eb-1c6d-4619-9964-fb853682d581.pdf. thumbPath: E:\k9.istructe.org\CMS\webtest\files\pdfthumbs\dd6fb1eb-1c6d-4619-9964-fb853682d581_1.png. objDoc: 1 - True. objPreview.Log: . strFileName: dd6fb1eb-1c6d-4619-9964-fb853682d581_1.png

Members/subscribers must be logged in to view this article

Correspondence on An Introduction to Neural Computing for the Structural Engineer by Professor W.M. Jenkins

Dr J. W. Smith (F)/Mr S. M. Williams (University of Bristol) The author poses the question, ‘is there anything in neural computing for the structural engineer?’ When the finite element method was first proposed its potential in structural engineering was immediately recognised. In contrast, neural computing has certain fundamental attractions and yet its application in structural engineering is not immediately obvious. The author concentrates on the principles of the method but does not provide an example of clear benefit over other methods. The writers would like to suggest reasons why neural computing has been slow to establish itself in structural engineering, and will suggest an application that may be tackled much more effectively by neural computing than by other methods.