STUDY. OF STAGE-DISCHARGE RELATIONSHIP USING ARTIFICIAL NEURAL- NETWORK

Abstract

The management of any river valley project greatly depends on the accurate assessment of water availability so that its yield can be properly optimized. Therefore the measurement of discharge of any gauging site plays a paramount importance. The discharge records are studied for a variety of applications. We are very often interested in annual peak discharge with a given return period, maximum probable floods, low flow for drought management/ control and also for assessment of hydropower potential. The measurement of discharge is an expensive affair. Therefore the conventional technique is to obtain a continuous record of river , stage. From the observed stage and discharge, a rating curve correlating stage and discharge is established. The subsequent procedure consists of reading discharge (Q) corresponding to measured stage (G) from G-Q relation. One of major limitation in the rating curve is that it does not take into account the hysterisis effect . Secondly the proper assessment of volume of sediment transported by a river is of vital interest due to its importance in design and management of water resources project. The conventional method of quantifying the sediment is from sediment rating curve which is plot of sediment concentration versus discharge....