<![CDATA[The transition to using digital weather observation equipment has improved efficiency but has also introduced new challenges regarding sensor accuracy and data reliability. This study aims to develop a predictive model for air temperature (T), relative humidity (RH), and atmospheric pressure (P) sensor values by applying an Artificial Neural Network (ANN) with a backpropagation algorithm. The data used consist of one-minute interval digital weather observation data collected from the Fatmawati Soekarno Class III Meteorological Station in Bengkulu. The ANN model architecture includes two hidden layers optimized through training and validation processes. The model’s performance was evaluated using the correlation coefficient (R), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The results show that the air temperature and relative humidity models achieved a high correlation (R > 0.94), with low RMSE and MAPE values. Meanwhile, the atmospheric pressure model, although having a moderate correlation (R ~0.56), was able to produce highly precise predictions (MAPE ~0.09%). Testing of the predictive models confirmed that they consistently maintained good performance. These findings demonstrate that the ANN model can be effectively used to support operational real-time monitoring of weather conditions and enhance the reliability of digital weather observation data.]]>
Copyrights © 2025
