Rainfall Monthly Prediction Based on Artificial Neural Network: A Case Study in Tenggarong Station, East Kalimantan - Indonesia

Abstract

The accuracy of forecasting rainfall is very important due to the current world climate change. Afterwards, to get an accurate forecasting of rainfall, this paper applied an Artificial Neural Network (ANN) with the Backpropagation Neural Network (BPNN) algorithm. In this experiment, the rainfall data were tested using two-hidden layers of BPNN architectures with three different epochs which were [2-50-10-1, epoch 500]; [2-50-20-1, with epochs 1000 and 1500]. The mean square error (MSE) is employed to measure the performance of the classification task. The experimental results showed that the architecture [2-50-20- 1, epoch 1000] produced a good result with the value of MSE was 0.00096341. Furthermore, BPNN algorithm has provided a good model to predict rainfall in Tenggarong, East Kalimantan - Indonesia.