Rainfall in the sea is very difficult to predict. there are many factors that become parameters in determining the estimated rainfall in the sea, by taking the correlation between rainfall and sea level will be able to predict rainfall with particle swarm optimization, One such technique which can be conveniently used to determine the model parameters accurately. This robust optimization technique has already been applied to improve the performance of artificial neural networks for time series prediction. This study uses particle swarm optimization technique to determine the parameters of an exponential autoregressive model for time series prediction. The model is applied for annual rainfall prediction and it shows a fairly good performance in comparison to the statistical ARIMA model. The research was conducted in Semarang, using data from rainfall, precipitation, and sea level rise in 2016. Network training by using one unit of input layer, two hidden layer units, and one unit of output layer. The first hidden layer with 10 neurons and the second hidden layer used 5 neurons. The best results on the training and testing of the network by using the parameter learning rate 0.3 and a momentum 0.6. The results obtained in the training get a percentage value of correlation is 79.0% and in the testing process to get the percentage correlation is 77.5%.

27/11/2018

978-602-53524-0-9

© The authors.
This article is distributed under the terms of the Creative Commons Attribution License 4.0, which permits non-commercial use, distribution and reproduction in any medium, provided the original work is properly cited. See for details: https://creativecommons.org/licenses/by-nc/4.0/

388 times.