Water level modeling for Kelantan River at Jeti Kastam Station using nonlinear autoregressive with exogenous input structure
| Autor: | Nurul Shakila Ahmad Zubir, Mohd Hezri Fazalul Rahiman, Khairah Jaafar |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Nonlinear autoregressive exogenous model
Mean squared error Meteorology Artificial neural network Flood myth 05 social sciences 0211 other engineering and technologies 021107 urban & regional planning 02 engineering and technology Regression Water level Nonlinear system Autoregressive model 0502 economics and business Statistics 050203 business & management Mathematics |
| Zdroj: | 2016 7th IEEE Control and System Graduate Research Colloquium (ICSGRC). |
| DOI: | 10.1109/icsgrc.2016.7813317 |
| Popis: | Generally, overflowing and unexpected amount of water level from normal conditions, especially in areas that are usually dry it is called flood. Kelantan River was synonym with flood especially during the months of November to February because of the northeast monsoon season. Nonlinear autoregressive with exogenous input (NARX) is well-known as one of the technique that has the ability to predict with efficient and good performance. River at Jeti Kastam Station was used in this study to predict water level using NARX model. The selection of Neural Network structure for water level and rainfalls modelling in Jeti Kastam Station was optimized and also the training and testing were analyzed. The performance of network was evaluated using Mean Square Error (MSE). It is shown that seven number of neurons in five number of delay afforded the lowest MSE validation, 1.43. The Regression, R for validation network is closed to 1 (0.9406), supports that the model is acceptable and able in predicting water level at Jeti Kastam station. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|