Due to the inherent non-stationary and nonlinear characteristics of original streamflow and the complicated relationship between multi-scale predictors and streamflow, accurate and reliable monthly streamflow forecasting is quite difficult. In this paper, a multi-scale-variables-driven streamflow forecasting (MVDSF) framework was proposed to improve the runoff forecasting accuracy and provide more information for decision-making. This framework was realized by integrating random forest (RF) and Gaussian process regression (GPR) with multi-scale variables (hydrometeorological and climate predictors) as inputs and is referred to as RF-GPR-MV. To validate the effectiveness and superiority of the RF-GPR-MV model, it was implemented for multi-step-ahead monthly streamflow forecasts with horizons of 1 to 12 months for two key hydrological stations in the Jinsha River basin, Southwest China. Other MVDSF models based on the Pearson correlation coefficient (PCC) and GPR with/without multi-scale variables or the PCC and a backpropagation neural network (BP) or general regression neural network (GRNN), with only previous streamflow and precipitation, namely, PCC-GPR-MV, PCC-GPR-QP, PCC-BP-QP, and PCC-GRNN-QP, respectively, were selected as benchmarks. Experimental results indicated that the proposed model was superior to the other benchmark models in terms of the Nash–Sutcliffe efficiency (NSE) for almost all forecasting scenarios, especially for forecasting with longer lead times. Additionally, the results also confirmed that the addition of large-scale climate and circulation factors was beneficial for promoting the streamflow forecasting ability, with an average contribution rate of about 15%. The RF in the MVDSF framework improved the forecasting performance, with an average contribution rate of about 25%. This improvement was more pronounced when the lead time exceeded 3 months. Moreover, the proposed model could also provide prediction intervals (PIs) to characterize forecast uncertainty, as supplementary information to further help decision makers in relevant departments to avoid risks in water resources management.

Wind speed and streamflow series always are nonlinear and unstable because the effects of chaotic weather systems. These inherent features make them difficult to forecast, especially in a changing environment. To improve forecasting accuracy, an innovation uncertainty forecasting architecture is developed by coupling data decomposition method, feature selection, multiple artificial intelligence (AI) techniques and composite strategy to do unstable time series forecasting. In the designed architecture, the AVMD (adaptive variational mode decomposition) is first applied to excavate implicit information from the original time series. Then, the random forest is utilized to select the suitable inputs for each mode. After that, the GPR (Gaussian Process Regression), a very famous probabilistic AI technique, is driven by various neural networks (ELM (Extreme Learning Machine), BP (Back Propagation Neural Networks), GRNN(Generalized Regression Neural Networks) and RBF (Radial Basis Function Neural Networks)) to produce both deterministic and probabilistic forecasting results in a nonlinear manner to play strengths of each other. The effectiveness and applicability of the proposed approach is verified by unstable wind speed data and streamflow data, and also compared with eleven related models. Results indicate that the proposed model not only improves the forecasting accuracy for deterministic predictions, but also provides more probabilistic information for decision making. The proposed method achieves significantly better performance than the traditional forecasting models both on wind speed forecasting and streamflow forecasting with at least 50% average performance promotion over all the eleven competitors. Comprehensive comparisons demonstrate the superior performance of the proposed method than the involved models as a powerful tool for unstable series forecasting.

This paper investigates the nonlinear modeling and stability of a doubly-fed variable speed pumped storage power station (DFVSPSPS). Firstly, the mathematical model of DFVSPSPS with surge tank considering nonlinear pump turbine characteristics was derived and established. Then, Hopf bifurcation analysis of DFVSPSPS was performed. The stable region was identified and verified by example analysis. Moreover, the effect mechanism of nonlinear pump turbine characteristics on the stability of DFVSPSPS was explored. Finally, the influence of factors on the stability and dynamic response of DFVSPSPS was studied. The results indicate that the emerged Hopf bifurcation of DFVSPSPS is supercritical and the region on the low side of the bifurcation line is the stable region. Nonlinear head characteristics have a significant influence on the stability and dynamic response of DFVSPSPS. Nonlinear speed characteristics have an obvious effect on the stability and dynamic response of DFVSPSPS only under positive load disturbance and unstable surge tank. Nonlinear head characteristics are unfavorable for the stability of DFVSPSPS under positive load disturbance and favorable under negative load disturbance. A smaller flow inertia of penstock, a smaller head loss of penstock and a greater unit inertia time constant are favorable for the stability of DFVSPSPS. The stable region under the positive disturbance of active power is larger than that under the negative disturbance of active power. The time constant of the surge tank presents a saturation characteristic on the stability of DFVSPSPS.

The precise forecast of solar radiation is exceptionally imperative for the steady operation and logical administration of a photovoltaic control plant. This study proposes a hybrid framework (CBP) based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), an enhanced Gaussian process regression with a newly designed physical-based combined kernel function (PGPR), and the backtracking search optimization algorithm (BSA) for solar radiation forecasting. In the CEEMDAN-BSA-PGPR (CBP) model, (1) the CEEMDAN is executed to divide the raw solar radiation into a few sub-modes; (2) PACF (partial autocorrelation coefficient function) is carried out to pick the appropriate input variables; (3) PGPR is constructed to predict each subcomponent, respectively, with hyperparameters optimized by BSA; (4) the final forecasting result is produced by combining the forecasted sub-modes. Four hourly solar radiation datasets of Australia are introduced for comprehensive analysis and several models available in the literature are established for multi-step ahead prediction to demonstrate the superiority of the CBP model. Comprehensive comparisons with the other nine models reveal the efficacy of the CBP model and the superb impact of CEEMDAN blended with the BSA, respectively. The CBP model can produce more precise results compared with the involved models for all cases using different datasets and prediction horizons. Moreover, the CBP model is less complicated to set up and affords extra decision-making information regarding forecasting uncertainty.