A Multivariate and Multistage Medium- and Long-Term Streamflow Prediction Based on an Ensemble of Signal Decomposition Techniques with a Deep Learning Network

Table 1

Selection of the input variables for different models.

Models	Input variables	Target variable
RBF		One-month ahead streamflow
SVR
RFR
GRU
LSTM
ICEEMDAN-LSTM	, Q (IMF1–IMF7, Residual), T (IMF1–IMF7, Residual), and (IMF1–IMF9, Residual)
VMD-LSTM	, Q (VF1–VF7, Residual), T (VF1–VF7, Residual), and (VF1–VF9, Residual)
IVL (ICEEMDAN-VMD-LSTM)	, Q (IMF2–IMF7, Residual), T (IMF2–IMF7, Residual), (IMF2–IMF9, Residual), QIMF1 (QF1–QF8), TIMF1 (TF1–TF8), and PIMF1 (PF1–PF10)