PREDICTION RAINFALL AND RAINY HOUR WITH ADAPTIVE NEURAL FUZZY INFERENCE SYSTEM (A N F I S) (Case Study: PT. Adaro, Kalimantan Selatan)

• Main Author: RICARDO LUMBAN GAOL (NIM: 12808037) , RICO
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/20450
• Institution: Institut Teknologi Bandung

The prediction is done as the parameter of the rainfall and the amount of rainy hour in South Borneo especially in the area of PT Adaro Indonesia. Prediction technique is done through training that the result of prediction could be varied for each member function of the prediction. In the other word, it can give uncertainty value of the prediction result. The prediction results of the rainfall produce RMSE (Root Mean Square Error). Temporary model as the result of the A N F I S identification can be used to predict the rainfall and the rain hour. Based on the verification result of the rainfall and rainy hour prediction, the correlation value which are 96% and 94% is finally collected. Scenario 3 is with 3 input CH(t-6); CH(t-12); CH(t-18). It is expected that the result of this study could be a consideration in later prediction system. Also, it can be concluded that A N F I S method can be used accurately as the method of rainfall and rainy hour prediction in the area.