Multiple Input Single Output (MISO) ARX and ARMAX model of flood prediction system: Case study Pahang

Year end usually would see many states in Malaysia would probably hits by severe floods due to Monsoon rain especially in the east coast. Many peoples suffered properties damages and economic losses. Thus, an accurate flood water level prediction model is required as an alarm system that would warn the affected area and residence to prepare for evacuation due to the upcoming severe flood. This paper compared the prediction performances of a developed flood prediction models that were designed using Multiple-Input Single-Output (MISO) Auto regressive with Exogenous Input (ARX) and MISO Auto regressive Moving Average with Exogenous Input (ARMAX) structure. The models were designed using Matlab System Identification toolbox of parametric model. The location for the case of study was at Pahang River, Temerloh, Pahang with four upstream stations and one downstream station or observed location. The data used were obtained from the Malaysian Department of Drainage and Irrigation. Simulation results showed that the prediction performance of flood prediction model designed by ARMAX structure showed better Best Fit value and smaller rmse values as compared to the model designed using ARX structure.