Naslov
A NEURAL NETWORK MODEL FOR RAINFALL-RUNOFF IN URBAN AREA

Autori
Šperac, Marija ; Varevac, Damir

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the XXIInd Conference of Danubian Countries on the Hydrological Bases of Water Management / - Brno, 2004

ISBN
80-86690-19-9

Skup
XXIInd Conference of Danubian Countries on the Hydrological Bases of Water Management

Mjesto i datum
Brno, Češka Republika, 30.08.2004. - 02.09.2004

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Artificial Nerual Networks; rainfall-runoff; urban area; hydrological inflow estimator; sewage system

Sažetak
The rainfall-runoff process is believed to be highly nonlinear, time varying, spatially distributed, and not easily described by simple models. The process consist of the movement of rainfall through different media and its transformation to the runoff in channels either natural or man made. Run-off in an urban setting differs from the one in a natural setting. In built-up settings the shape of a run-off hydrogram is changed, the time of concentration is decreased, the peak of hydrogram is increased, a smaller amount of water is infiltrated into the underground, a volume of surface waters is increased up to two times. Run-off in an urban catchments area presents a complex problem, which calls for the integral approach. Precipitation and run-off as basic parameters of a hydrologic cycles are of a stochastic character. An Artificial Neural Networks is a flexible mathematical structure, which is capable of identifying complex nonlinear relationship between input and output data sets. The results of applicability of neural network for determination of run-off in the urban discharge system show advantage in comparison to standard methods. These advantages particularly refer to the possibility of fast and easy generation of hypothetical hydrograms of runoff in sewage systems. Such hydrograms can be applied in the procedures for finding best management decisions when solving tasks related to operational management of sewage systems.

Izvorni jezik
Engleski

Znanstvena područja
Građevinarstvo

POVEZANOST RADA