The intermittency and volatility of wind power (WP) have restricted the large-scale integration of wind turbines into power systems. Therefore, many researches on the deterministic and probabilistic forecasting of WP are being more important for power system planning and operation. In this work, a deterministic and probabilistic interval prediction for wind farm based on variational mode decomposition (VMD) and weighted least squares support vector machine (LS-SVM) were suggested. VMD is proposed to handle the variability of the novel WP series. In order to overcome the influence on outliers and non-Gaussian error distributions, a weighted LS-SVM is adopted to build deterministic prediction model for WP. Besides, lower upper bound estimation (LUBE) method for probabilistic interval forecasting is presented to quantify the potential risks of the WP series. The LUBE of the optimal prediction intervals is calculated. In the comparative experiments, prediction intervals coverage probability (PICP), prediction intervals normalized average width (PINAW) and normalized average deviation (NAD) are demonstrated to appraise the probabilistic prediction of WP. The simulation results show that the proposed method has much greater performance than other methods.

Wind power forecasting is essential for maintaining grid stability due to the inherent intermittency and nonlinear dynamics of wind energy. This study proposes a robust two-stage hybrid framework for short-term wind power prediction integrating variational mode decomposition (VMD), artificial neural networks (ANNs), and weighted least-squares support vector machine (W-LSSVM). In the first stage, VMD decomposes wind speed time series into band-limited intrinsic mode components to reduce non-stationarity and noise. Each component is predicted using a nonlinear ANN model. In the second stage, a residual-based adaptive weighted LSSVM aggregates the predicted components to estimate SCADA-measured wind power, mitigating the influence of outliers and noisy samples. A chronological rolling-origin evaluation protocol is employed to preserve temporal ordering and assess generalization within the evaluated temporal windows. Performance is evaluated over multiple non-overlapping test windows and reported as mean ± standard deviation. Comparative experiments against classical baselines, persistence forecasting, and LSTM models demonstrate competitive short-horizon accuracy and improved robustness under the evaluated degraded-data conditions. Empirical prediction intervals additionally provide an assessment of forecast uncertainty. The results indicate that the proposed framework provides competitive short-term forecasting performance and improved robustness under the evaluated degraded-data conditions. However, validation using longer-term and multi-site datasets is required to establish broader generalizability.