The Discharge Forecasting of Multiple Monitoring Station for Humber River by Hybrid LSTM Models

An early warning flood forecasting system that uses machine-learning models can be utilized for saving lives from floods, which are now exacerbated due to climate change. Flood is carried out by determining the river discharge and water level using hydrologic at target sites. If forecasted reach dangerous levels, sends messages residents in flood-prone areas. In past, hybrid Long Short-Term Memory (LSTM) have been successfully used time series forecasting. However, prediction errors grow exponentially with period, making forecast unreliable as an tool enough lead time. Therefore, this research aimed improve accuracy of employing real-time monitoring network datasets establishing temporal spatial links between adjacent stations. We evaluated performance LSTM, Convolutional Neural Networks LSTM (CNN-LSTM), (ConvLSTM), Spatio-Temporal Attention (STA-LSTM) The dataset, employed validation, includes hourly records, 2012 2017, on six stations Humber River City Toronto, Canada. Experiments included both 6 12 h ahead, data input past 24 h. STA-LSTM model’s was superior CNN-LSTM, ConvLSTM, basic when longer than