Accurate wind power forecasting is critical for enhancing the operational efficiency and stability of electrical power grids. Conventional single-variable signal decomposition forecasting methods ignore the coupling relationship between wind power and multiple meteorological data, thus limiting prediction accuracy. This study proposes an accurate and fast short-term wind power prediction approach based on series-core fusion technology considering multiple meteorological data. In the data preprocessing stage, the multivariate variational mode decomposition (MVMD) algorithm decomposes wind power and meteorological variables into the same predefined number of frequency-aligned intrinsic mode functions (IMFs), thereby enhancing feature representation and improving forecasting accuracy via a more comprehensive and detailed dataset representation. During the training stage, the series-core fused time series (SOFTS) model establishes the connection among wind power channel and other meteorological variable channels for each IMF, achieving fast convergence through its streamlined and parallel structure. In the forecasting stage, the final wind power prediction is generated by the reconstruction of all IMFs. Furthermore, we conducted a comprehensive performance evaluation by comparing the proposed MVMD-SOFTS model with eight alternative models, including the CNN model, the TCN model, the LSTM model, the GRU model, the Transformer model, the SOFTS model, the CEEMDAN-SOFTS model, and the VMD-SOFTS model. The results indicate that MVMD-SOFTS outperformed all other models, demonstrating its effectiveness in capturing the multifaceted relationships in wind power forecasting.

To address the limitations of existing power system low-carbon dispatching studies—such as over-reliance on generation-side carbon mitigation, price-oriented demand response (DR) failing to guide carbon reduction, and the low solution efficiency of traditional carbon emission flow (CEF)-based two-stage models—this paper proposes a data-driven CEF framework integrated with a bi-level economic and low-carbon dispatching model. First, a data-driven CEF calculation method is developed: It eliminates the need for complex power flow post-processing while maintaining calculation accuracy through multiple linear regression. On this basis, a bi-level optimization model is constructed: The upper level focuses on optimizing the economic and low-carbon objectives of power grid operation, while the lower level regulates industrial, commercial, and residential load aggregators (LAs) via carbon-intensity-oriented DR strategies and economic compensation mechanisms. Finally, a sample-based optimization algorithm combined with convex relaxation is proposed to solve the model, avoid the static setting of power flow and carbon intensity, and improve solution efficiency. Case studies demonstrate the following: the proposed method reduces the calculation time of node carbon intensity from 5 min to less than 100 ms, with the coefficient of determination (R²) ranging from 0.969 to 0.998; compared with the two-stage method, it achieves a 4.26% reduction in total scheduling cost, a 3.80% decrease in total carbon emissions, a 53.27% drop in carbon trading cost, and a 21.6% shortening in iteration time. These results verify that the proposed method can effectively enhance the source−load interaction and improve the accuracy and efficiency of low-carbon scheduling. This study provides a feasible technical path for the low-carbon transition of new-type power systems.