Analysis of the Uncertainty of the Tipping Bucket Type of Rainfall Sensor Measurement based on the Internet of Things
Rain is very difficult to predict. This is because there ar eso many factors that can affect rain starting from temperature, humidity, rainfall and intensity of sunlight. Moreover, coupled with weather anomalies such as la nina and el nino which cause a longer rainy period than usual. Whereas high rainfall causes disasters such as floods and so on. Therefore, it is important in predicting the rain that will occur in a place As it is likely possible anticipate flood disaster that will occur. This study uses a backpropagation type of artificial neural network in predicting rainfall. Input data that used to train this artificial neural network is data from BKMG about monthly rainfall during 2015-2019. Based About the result of the conducted test data, the MSE at output of the artificial neural network is 0.089161. From these effects it could be assume that the synthetic neural network with method backpropagation works well to predict the rainfall that will occur.
Keyword : Backpropagation, Rainfall, Prediction, Artificial Neural Network.