PREDICTION OF PORE WATER-PRESSURE USING GENERALIZED REGRESSION NEURAL NETWORK

In the decade, the prediction and forecasting of hydrologic time series had been nominated by Artificial Neural Network (ANN) because their ability in practical application although numerous model has been established for prediction and development in various engineering fields.
In this paper, the study was proposed to predict pore water pressure by using Generalized Regression Neural Network with these following detailed objectives which is to predict pore water pressure by using Generalized Regression Neural Network and then evaluated the performance of the model using root mean square error (RMSE), coefficient of determination (R2) and coefficient of efficiency (CE). Intention to perform prediction by using GRNN were because the study was never been conducted yet by using time series pore water pressure response to rainfall. Moreover, the previous researchers were suggested the further study should be conducted in order to get a better result in prediction of pore water pressure. They described the study by using estimation was hard and truly challenging to reach a preferred and accurate result if only had a few study be conducted.
In methodology part was about the procedure how to develop model from data. The steps start with slope instrumentation for getting the data, continued with method application on how to perform the data collection, selection, processing, separation and analysis. Then, MATLAB application and model development for prediction method was carry out to select of model performance