Accurate short-term wind power forecasting is essential for the secure and economic operation of power systems with high renewable energy penetration. However, forecasting performance is still affected by meteorological forecast uncertainty and the complex multi-scale fluctuation characteristics of wind power generation. To address these challenges, this paper proposes an Initial-State-Aware Multi-Scale Transformer framework for 12 h-ahead wind power forecasting. The principal methodological contribution is an initial-state-aware meteorological representation and progressive fusion strategy tailored to weather-driven wind power forecasting. The framework explicitly distinguishes the atmospheric state available at forecast initialization from the subsequent forecast meteorological trajectory and uses the former to condition the representation of the latter through cross-attention and gated residual fusion. The resulting meteorological representation is then progressively coupled with coarse- and fine-scale historical power representations, and a horizon-oriented forecasting head generates the future power sequence in parallel. Experiments on three wind farms demonstrate that the proposed method achieves the best overall forecasting performance. Compared with the strongest baseline model, it reduces NMAE and NRMSE by 8.54% and 2.36%, respectively.
Chao-Ying Yang, Jun Zhao, Peng Han et al.· Energies· 0 citations
High renewable penetration makes day-ahead operation sensitive to the directional effects of wind and photovoltaic forecast errors. Conventional prediction intervals mainly evaluate coverage and sharpness, but lower- and upper-boundary violations correspond to different operational risks: shortage-side supply-adequacy pressure and accommodation-side curtailment pressure. This paper proposes context-aware asymmetric conformal quantile regression (CA-ACQR) to construct directional renewable-power risk intervals. The method builds separate conformal scores for the two interval sides, estimates context-dependent boundary corrections, and reallocates the tail-risk budget under supply-priority, balanced, and accommodation-priority profiles. Case studies use regional wind and photovoltaic power data, with contextual groups defined by renewable type, lead-time block, forecast difficulty, weather-risk regime, and output level. CA-ACQR increases the prediction interval coverage probability (PICP) from 91.11% to 94.07% and reduces the accommodation-side violation rate from 4.37% to 1.44%. The results demonstrate selectable directional risk postures and quantify trade-offs among interval width, directional violations, normalized stress cost, and the 95% conditional value-at-risk stress cost.
Peng Han, Jun Zhao, Yu Liu et al.· Energies· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
