engleski

Mošćenička Draga Early Warning Systems Development Using Machine Learning

The paper presents a machine learning model for predicting Mošćenička Draga torrential stream water levels and discharges based on meteorological and hydrological data. Possible use of machine learning algorithms for early warning system development is presented as well. One of the most important goals of Croatian – Japanese Project Risk Identification and Land- Use Planning for Disaster Mitigation of Landslides and Floods is development of the Early Warning Systems (EWS). The most important task of the EWS in Mošćenička Draga is identification of the critical torrential discharges and precipitations that can cause floods, early enough to inform the authorities and public. The paper describes an artificial intelligence model that has been developed by integrating meteorological and hydrological data. The WEKA Data Mining Software was used for a model development. Model has been validated on measured hydrological and meteorological data from May 2011 till December 2012. Water levels predicted by the model are in accordance with the measured data. Model validation showed that the mac

early warning system; machine learning; flood; Mošćenička Draga

nije evidentirano