Application of Artificial Neural Networks using the Matlab Application for Predicting the Rate of Development of Rainfall Intensity (Case Study in Rangkasbitung District)

Abstract

With the development of increasingly advanced information technology at this time, the MatLab application program created by applying an Artificial Neural Network can be used to predict the rate of development of rainfall intensity. Information technology can now be used to assist in processing, obtaining, compiling, storing, and managing data to produce information for researchers in compiling research so that research becomes very easy. Food crop production is influenced by rainfall during the development and growth of food crops; this causes rainfall prediction to be essential in agricultural planning. Prediction is the most crucial tool in determining everything effectively and efficiently. Therefore, several methodologies are used in compiling this research, namely collecting rainfall intensity data from each year, checking each problem, and then matching it with the required data. Samples were taken to determine the criteria for rainfall intensity based on BPS (Central Statistics Agency) data, namely data on rainfall intensity in Rangkasbitung sub-district from 2010 to 2020. The data is presented in tabular form. The following is a table of rainfall intensity data. 1) The backpropagation algorithm can be used in the prediction process; whether it is good or not is influenced by a learning material and the number of neurons in the hidden layer. 2) If the data is getting more significant on the unit in the hidden layer, the prediction results will be closer to the targeted value. 3) Using an artificial neural network in the MatLab software is proven to predict rainfall intensity development rate.