River water level forecasting for flood warning system using deep learning long short-term memory network

Flood is considered chaotic, complex, volatile, and dynamics. Undoubtedly, its prediction is one of the most challenging tasks in time-series forecasting. Long short-term memory (LSTM) networks are a state of the art technique for time-series sequence learning. They are less commonly applied to the hydrological engineering area, especially for river water level time-series data for flood warning and forecasting systems. Yet it is inherently suitable for this subject. This paper examines an LSTM network for forecasting the river water level in Klang river basin, Malaysia. The river water level contains of single time series observed data, with time steps corresponding to hourly data and values corresponding to the water level or stage level in meters. In this study, prediction responses for river water level data using a trained recurrent neural network and update the network state function is applied. The result verified that the LSTM network with specified training set options is a promising alternative technique to the solution of flood modelling and forecasting problems. The performance indicates with the root mean square error, RMSE 0.20593 and coefficient of determination value, R2 0.844 are closely accurate when updating the network state compared with the observed values.