Građevinar, Vol. 61 No. 07., 2009.
Preliminary communication
Estimating deflection of cable stay bridges by support vector method
Dean Čizmar
Vlatka Rajčić
Anđelko Vlašić
Abstract
Authors present a new artificial-intelligence method called the support vector method. The method is used in the classification of samples and for regression analysis. The set of data defining rigidity parameters for cable stay bridges has been selected as an example. When used in the scope of this method, the mentioned data enable estimation of cable stay bridge deflection due to traffic load. The results obtained are compared with real deflection values, and also with deflections obtained by neural network. It is claimed that the method is highly accurate for estimating deflection.
Keywords
cable stay bridge; deflection; artificial intelligence; support vector method; traffic load; neural network
Hrčak ID:
40585
URI
Publication date:
3.8.2009.
Visits: 1.789 *