Hydrological peaks evaluation at Chitral and Tarbela stations, Pakistan, using combined Bayesian regularized neural network and signal difference average based variational mode decomposition method: A case study

Abstract

Pakistan being an agricultural country highly depends on its natural water resources originate from the upper regions of Hindu Kush-Karakoram-Himalaya Mountains and nourish one of the world's largest Indus Basin irrigation system. This paper presents streamflow modelling and forecasting using signal difference average (SDA) based variational mode decomposition (VMD) combined with machine learning (ML) methods at Chitral and Tarbela stations on the Indus River network. For this purpose, VMD based; random forest (VMD-RF), gradient boosting machine (VMD-GBM) and Bayesian regularized neural network (VMD-BRNN) have been opted. Moreover, traditional time series flow model that is seasonal autoregressive integrated moving average (SARIMA) and classical decomposition approach with particle swarm optimization-based support vector regression (PSO-SVR) are considered as benchmark models for comparison. The results show that overall, VMD-BRNN performed best, followed by VMD-GBM and VMD-RF, whereas, SARIMA and PSO-SVR ranked last. Overall, SARIMA and PSO-SVR are failing to capture most of the peaks even during training period. Whereas, hybridization of VMD and ML methods has shown increased robustness of the models. The results show that the influential role of the high dimensional components and robustness on the river flow may be explored by most optimum SDA based VMD signals hybrid with BRNN method.

Prediction of Landslide Displacement Based on the Combined VMD-Stacked LSTM-TAR Model

“…Conventional decomposition methods require decomposition of modeling samples, which reduces the subsequent training data. The variational mode decomposition (VMD) method can decompose the cumulative landslide displacement as a signal in the frequency domain and alleviate the modal aliasing problem in the common empirical mode decomposition method [14][15][16]. For the fluctuation displacement series after separating the overall trend, the statistical machine learning models were normally used, such as the back propagation neural network (BPNN) [17] and extreme machine learning [18]; however, they are all static network models, and their ability to process changes in a time series is weak.…”