Wavelet-ANN model for flood events

Abstract

The observation of peak flows into river or stream system is not straight forward but complex function of hydrology and geology of the region. There are well established statistical approach to predict the flood events with their magnitude and frequency. Development of models based on temporal observations may improve understanding the underlying hydrological processes in such complex phenomena. Present work utilized temporal patterns extracted from temporal observations of annual peak series using wavelet theory. These patterns are then utilized by an artificial neural network (ANN). The wavelet-ANN conjunction model is then able to predict the flood event comparable to statistical approach. The application of the proposed methodology is illustrated with real data. The limited performance evaluation of the methodology show potential application of the developed methodology. © 2012 Springer India Pvt. Ltd.