The Neural Network Models which mathematically interpret human thought processes were applied to resolve the uncertainty of model parameters and to increase the model's output for the streamflow forecast model. In order to test and verify the flood discharge forecast model eight flood events observed at Kumho station located on the midstream of Kumho river were chosen. Six events of them were used as test data and two events for verification. In order to make an analysis the Levengerg-Marquart method was used to estimate the best parameter for the Neural Network model. The structure of the model was composed of five types of models by varying the number of hidden layers and the number of nodes of hidden layers. Moreover, a logarithmic-sigmoid varying function was used in first and second hidden layers, and a linear function was used for the output. As a result of applying Neural Networks models for the five models, the N10-6model was considered suitable when there is one hidden layer, and the N10-9-5model when there are two hidden layers. In addition, when all the Neural Network models were reviewed, the N10-9-5model, which has two hidden layers, gave the most preferable results in an actual hydro-event.

This study aims at the effective estimation of water supply capacity of small scale reservoir and the proposal of the data which is necessary to establish the water resources management plan of down stream area of the reservoir in the future by comparison and examination about reservoir operation technique for the security of agricultural water in small scale reservoir. The result of flow calculation by Tank model is used for the input data as the inflow data which is needed for the analysis of water supply capacity. Stochastic method, simulation method, and optimization method are used to examine the water supply capacity, and water security amount is compared with each method. From the analyses of water supply capacities by each method, slightly different results are shown in spite of the effort to compare them equally using input data such as inflow data under equal conditions, and the comparison of water supply capacities by each method are as follows; linear planning method, simulation method, and transition probability matrix method in the order of amount from the largest. It is thought that the simulation method in which comparatively reasonable application of the inflow data is possible and is simulated in successive time series dam operation of the three methods used in this study thus, simulation model is proper to estimate the water supply capacity of agricultural small scale reservoir. And it is judged that the heightening of efficiency of water resources utilization according to the development of downstream area of dam may be possible using the upward readjusted water supply amount of 55.18×106ton and 63.7×106ton at 95% and 90% supply reliability respectively which are above the planning water supply amount of 50.0×106ton when the simulation method is introduced as the standard.