PERAMALAN CURAH HUJAN DI KALIMATAN SELATAN DENGAN JARINGAN SYARAF TIRUAN

Abstract

South Kalimantan is in the area of high rainfall so it is included in the criteria of the rainy season. Artificial Neural Network (ANN) is one method that can identify patterns of data from rainfall forecasting system by conducting training method. One of the model ANN used is Backpropagation (BP). The training of a network using BP consists of 3 steps, namely: feedforward input pattern training, calculation and BP from the set of error and weight adjustment. The purpose of this research is to predict rainfall in South Kalimantan in 2015 using JST BP. The research method used in this research is literature study and case study related to rainfall forecasting, JST and BP. This research procedure will begin by collecting data, analyzing data and training data then predicting the data to be achieved. The results of this research is the highest rainfall in South Kalimantan in 2015 occurred in the area of Martapura Kota Kab. Banjar in January. In this period of the month there is a possibility that the area will experience an increase in water level or flood. While the lowest rainfall occurred in the region Pelaihari Kab. Land of the Sea around August and September. In this period the rainfall is so low that the area is likely to be in dry conditions.