Climate change is a long-term change in the ordinary weather conditions that the local, regional and global climates and one of the solutions to overcome this phenomenon is by managing water resources efficiently. Reservoir is a major player to achieve water resource management therefore it requires an accurate forecasting. SVR and MLPNN are introduced as a solution to achieve an efficient reservoir inflow forecasting, there are so many input parameters that influence reservoir water flow and the 3 most important parameter are storage level, rainfall, and finally evaporation that are being fed to the two models. There have been various parameters tested such as kernel types in SVR and number of hidden layers and neurons in MLPNN both models have proven its ability how-ever MLPNN with two hidden layers and 4 neurons in each layer outperformed SVR after being tested using four different tests.
Forecasting of Reservoir Inflow Using Machine Learning Case Study: Klang Gate Dam Reservoir