Ensemble streamflow forecasting based on variational mode decomposition and long short term memory

Abstract

Reliable and accurate streamflow forecasting plays a vital role in the optimal management of water resources. To improve the stability and accuracy of streamflow forecasting, a hybrid decomposition-ensemble model named VMD-LSTM-GBRT, which is sensitive to sampling, noise and long historical changes of streamflow, was established. The variational mode decomposition (VMD) algorithm was first applied to extract features, which were then learned by several long short-term memory (LSTM) networks. Simultaneously, an ensemble tree, a gradient boosting tree for regression (GBRT), was trained to model the relationships between the extracted features and the original streamflow. The outputs of these LSTMs were finally reconstructed by the GBRT model to obtain the forecasting streamflow results. A historical daily streamflow series (from 1/1/1997 to 31/12/2014) for Yangxian station, Han River, China, was investigated by the proposed model. VMD-LSTM-GBRT was compared with respect to three aspects: (1) feature extraction algorithm; ensemble empirical mode decomposition (EEMD) was used. (2) Feature learning techniques; deep neural networks (DNNs) and support vector machines for regression (SVRs) were exploited. (3) Ensemble strategy; the summation strategy was used. The results indicate that the VMD-LSTM-GBRT model overwhelms all other peer models in terms of the root mean square error (RMSE = 36.3692), determination coefficient (R2 = 0.9890), mean absolute error (MAE = 9.5246) and peak percentage threshold statistics (PPTS(5) = 0.0391%). The addressed approach based on the memory of long historical changes with deep feature representations had good stability and high prediction precision.

Short-term inflow forecasting in a dam-regulated river in Southwest Norway using causal variational mode decomposition

“…Since the range of inflow and other Modes decomposed from meteorological and hydrological data vary widely, it is necessary to normalize all input data 22 . In this paper, all the variables are normalized to the same scale by using the MinMaxScaler package from scikit-learn 26 from range to 1.…”

“…The variable is the Dirac distribution, t is time script, and denotes convolution 21 .The optimization problem in Eq. ( 1 ), can be solved by turning it to a Lagrangian form and adding a quadratic penalty term to render the problem unconstrained as follows 22 : where is the Lagrangian function, is the dual variables, and indicates the balancing parameter of the data-fidelity constraint. The Alternate Direction Method of Multipliers (ANMM) is employed in VMD to solve the Lagrangian problem presented in Eq.…”

Post-Processing Ensemble Precipitation Forecasts and Their Applications in Summer Streamflow Prediction over a Mountain River Basin

“…Reliable and accurate streamflow prediction, along with its inherent uncertainty, is significant to water resource management and disaster mitigation [1][2][3][4]. Precipitation plays a vital role in streamflow prediction as a critical input for hydrological modeling.…”

Bias Correction for ERA5-Land Soil Moisture Product Using Variational Mode Decomposition in the Permafrost Region of Qinghai–Tibet Plateau

“…Variational mode decomposition (VMD) is a novel mode decomposition method which is based on a strict mathematical foundation, overcoming the problems of end-point effect, mode-mixing phenomenon, and noise sensitivity in the class of EMD methods [32], [33]. The application of VMD in the hydrological data series mainly includes river runoff prediction and trend analysis [34], [35].…”