Developing a novel layer network structure for a LSTM model to predict mean monthly river streamflow

İngilizce (OpenAlex)

In this research, novel innovative DDN layer network structures by hybridizing double-LSTM model with an addition layer (+) ( i.e., 2LSTM and 2LSTM + layer network models) are developed purposefully to enhance prediction performance of the mean monthly Maroon River streamflow ( MRSF m ) in Iran from October 1987 to September 2017. For doing so, to select the most effective parameters on MRSF m , the Pearson’s correlation coefficient (PCC) and Cosine amplitude sensitivity (CAS) as features selection process are carried out for potential meteorological variables in the study area ( i.e., average monthly temperature ( T m ), evaporation ( ET m ), and precipitation ( P m )) and target ( MRSF m ). The results show that T m and ET m have an insignificant influence on MRSF m , thus, only P m is used as the most effective input variable in predicting MRSF m . Due to a well-balanced network model’s structural outline in the suggested novel hybrid 2LSTM + model, it accordingly yields to a suitable total learnable parameter ( TLP ) compared to ordinary standalone LSTM and GRU as the benchmark models developed in the similar meta-parameters. This model under the optimal meant meta-parameters tuned i.e., state activation functions ( SAF ) = tanh-softsign , numbers of hidden neurons ( NHN ) = 75, dropout rate ( P-rate ) = 0.5, performs best among the models with an R 2 of 0.68, NSE of 0.63, PBIAS of 41%, KGE of 0.79, and RMSE of 19.24 m 3 /s. Comparatively, a standard gated recurrent units (GRU) and LSTM as benchmark models using the optimal scenario generate the following results: R 2 are 0.57 and 0.67, NSE are 0.53 and 0.61, PBIAS are 109 and 49%, KGE are 0.63 and 0.79, and RMSE are 21.11 and 19.32 m 3 /s, respectively. Generally, in all models, in the equal NHN , rising P-rate value reduces convergence time.