/Reinforcing-a-Monitoring-System-of-a-Regulated-River-using-ML

Reinforcing-a-Monitoring-System-of-a-Regulated-River-using-ML. A multilevel intelligent flood forecast model using a combination of multi recurrent neural network (RNN) and regression models. Through comparisons with autoregressive integrated moving average (ARIMA), Gated Recurrent Unit (GRN), Long-Short Term Memory (LSTM), and BiLSTM.

Primary LanguageJupyter Notebook

Reinforcing-a-Monitoring-System-of-a-Regulated-River-using-ML

Traditional flood prediction methods, such as rain forecast and satellite cloud imaging, have low prediction accuracy and only provide short-term early warnings. As a result, they do not give authorities enough preparation time. To address these limitations, we developed a multilevel intelligent flood forecast model using a combination of multi recurrent neural network (RNN) and regression models. Through comparisons with autoregressive integrated moving average (ARIMA), Gated Recurrent Unit (GRN), Long-Short Term Memory (LSTM), and dual-cycle LSTM (BiLSTM) models, we selected the BiLSTM model as the prediction module due to its higher accuracy and lower data loss.

Paper Published in ICAI'23