In response to climate change, the proportion of renewable energy penetration is increasing daily. However, there is a lack of flexible energy transfer mechanisms. The optimization effect of low-carbon economic dispatch in a single park is limited. In the context of the sharing economy, this study proposes a research method for multi-park electricity sharing and benefit allocation based on carbon credit trading. Firstly, a framework for multi-park system operation is constructed, and an energy hub model is established for the electrical, cooling, and heating interconnections with various energy conversions. Secondly, a park carbon emission reduction trading model is established based on the carbon credit mechanism, further forming an optimal economic and environmental dispatch strategy for multi-park electricity sharing. Matlab+Gurobi is used for solving. Then, based on asymmetric Nash bargaining, the comprehensive contribution rate of each park is calculated by considering their energy contribution and carbon emission reduction contribution, thereby achieving a fair distribution of cooperation benefits among multiple parks. The results show that the proposed method can effectively reduce the overall operational cost of multiple parks and decrease carbon emissions, and the benefit allocation strategy used is fair and reasonable, effectively motivating the construction of new energy in parks and encouraging active participation in cooperative operations by all parks.
Short-term load forecasting plays a pivotal role in modern power system operations yet it remains challenging due to the complex spatiotemporal dependencies in load data. This paper proposes a dual-head attention residual network (DARNet) that significantly advances STLF through three key innovations: (1) a hybrid encoder combining 1D-CNN and GRU architectures to simultaneously capture the local load patterns and long-term temporal dependencies, achieving a 28% better locality awareness than that of conventional approaches; (2) a novel dual-head attention mechanism that dynamically models both the inter-temporal relationships and cross-variable dependencies, reducing the feature engineering requirements; and (3) an autocorrelation-adjusted recursive forecasting framework that cuts the multi-step prediction error accumulation by 33% compared to that with standard seq2seq models. Extensive experiments on real-world datasets from three Chinese cities demonstrate DARNet’s superior performance, outperforming six state-of-the-art benchmarks by 21–35% across all of the evaluation metrics (MAPE, SMAPE, MAE, and RRSE) while maintaining robust generalization across different geographical regions and prediction horizons.
Introduction: Novel cable traction power supply can eliminate phase splits and improve regenerative energy utilization in metro systems. However, the cable-based traction network has a relatively complex structure, making internal energy consumption and train-network line loss evaluation insufficiently studied. Methods: To quantify train-network line losses, an equivalent solid-circuit model of the traction power supply system is established using an external cascade strategy. Combined with actual train operation load modeling, the traction transformer output power and network losses are computed and compared under both traction and regenerative braking operating conditions. Results: The proposed modeling-and-calculation framework enables accurate evaluation of power output and line losses across operating modes, and effectively captures the electrical characteristics of the cable traction energy supply system. Discussion: The developed theoretical line-loss calculation approach provides a quantitative basis for energy-consumption assessment and operational analysis of cable traction networks, supporting performance evaluation and planning of metro traction power supply systems.