APPLICATION STUDY OF STORAGE FUNCTION MODEL & KALMAN FILTERING FOR..FLOOD SIMULATION

Abstract

This dissertation study has been taken up with the objectives ; (i) to provide a solution of nonlinear storage function model and use of a procedure to update parameters of storage function model on real time basis (Schmidt - Kalman Filter ) ; (ii) to conduct an application study in Cisadane basin ( Indonesia ) ; ( iii) to carry out sensitivity analysis of the flood simulation model with respect to runoff coefficient, channel roughness and permissible error limits. This study uses a nonlinear storage function model converted from the kinematic wave equations. Kinematic routing parameters are correlated to storage function parameters. The study demonstrates that the nonlinear storage function model is a simple and convenient tool for flood simulation on real time basis. Two storm events (March 1981, February 1985 ) and related flood hydrographs for the Cisadane river have been used in comparing the observed and forecasted flood hydrographs ( with different lead times ). Comparison has been done in terms of ( i ) observed peak flood and forecasted peak flood and (ii) observed time to peak and forecasted time to peak. Observed and forecasted peak floods differ upto 100 cumecs . Observed and forecasted time to peak are more are less similar....