Modeling flood estimation using fuzzy logic & artificial neural network

• Main Authors: Borujeni, Sattar Chavoshi, Sulaiman, Wan Nor Azmin, Abd Manaf, Latifah, Sulaiman, Md Nasir, Saghafian, Bahram
• Format: Conference or Workshop Item
• Language: English
• Published: 2009
• Online Access: http://psasir.upm.edu.my/id/eprint/17808/1/50.pdf; http://psasir.upm.edu.my/id/eprint/17808/
• Institution: Universiti Putra Malaysia

Estimates of flood discharge with various risks of exceedance are needed for a wide range of engineering problems: examples are culvert and bridge design and construction floods in major projects. At a site with a long record of measured floods, these estimates may be derived by statistical analysis of the flow series. Alternatively the storm magnitude of an appropriate duration, aerial coverage and return period may be estimated and converted into the flood of a given return period using a rainfall/runoff model such as the unit hydrograph. However, in cases where adequate rainfall or river flow records are not available at or near the site of interest, it is difficult for hydrologists and engineers to derive reliable flood estimates directly and regional studies can be useful. This is particularly true in the case of semi-arid areas, where, in general, flow records are scarce. The problem of assigning a flood risk to a particular flow value is one which has received considerable attention in the literature. The estimation of flood risk through the evaluation of a flood frequency distribution is complicated, however, by the lack of a sufficient temporal characterization of the underlying distribution of flood events. The inadequacies in the data availability necessitate the estimation of the flood risk associated with events which have a return period beyond the length of the historical record. Regional flood frequency analysis can be effective in substituting an increased spatial characterization of the data for an insufficient temporal characterization, although problems exist with the implementation of regional flood frequency analysis techniques.