[Objective] To address the challenge of declining accuracy in photovoltaic power prediction caused by rapidly changing cloud conditions， this paper proposes an ultra-short-term photovoltaic （PV） power prediction method based on the cloud-irradiance-power coupling mechanisms. [Methods] First， a two-stream convolutional network is constructed to extract spatial and motion features of clouds， while a Transformer encoder is employed to capture the temporal features of irradiance sequences. Then， cross-attention mechanism is applied to fuse these features for multi-step ahead prediction of surface irradiance. Finally， to overcome the limitations of purely data-driven approaches， this study constructs a photovoltaic power prediction model by using irradiance forecasts as inputs and incorporating physical constraints governing the surface irradiance-to-power output conversion. [Results] Experimental results show that the proposed irradiance prediction method outperforms baseline models under various weather conditions. Furthermore， using the predicted irradiance as input significantly improves PV power forecasting accuracy compared to methods that directly incorporate cloud information. The introduction of physical constraints further enhances PV power prediction performance and improves the model’s generalization capability. [Conclusions] This study demonstrates that deeply mining the impact of dynamic 3D cloud information on irradiance and incorporating physical prior knowledge between irradiance and PV power represents an effective way to enhance the accuracy and model generalization capability of ultra-short-term PV power prediction.

[Objective] To address the problems that when large-scale electrification-transportation coupling network （ETCN） encounters sudden resilience faults， traditional schemes have slow generation speed， are difficult to respond to dynamic information interaction in real time， and artificial intelligence algorithms are prone to cause safety accidents such as voltage over-limit due to the lack of security mechanisms in application， this paper proposes a resilience improvement strategy for ETCN based on safe deep reinforcement learning （SDRL）. [Methods] First， the paper establishes a two-stage electrification-transportation optimization framework： the first stage prioritizes the protection of high-value loads with minimum reconfiguration cost， while the second stage optimizes electric vehicle （EV） routing with minimum traffic dispatch cost. Second， a hierarchical decision-making model based on a modified Rainbow algorithm is designed. The upper layer outputs the action plan of the power grid interconnection switch and inputs the reconstructed power grid state to the lower layer. The lower layer integrates grid reconfiguration state with real-time traffic information to optimize EV routing selection， with the objective to ensure that EV routing optimization can real-time adapt to the power grid’s recovery needs. In addition， the Lagrange multiplier safety mechanism is embedded， and an objective function with risk penalties is constructed to achieve dynamic penalties for risk behaviors such as voltage over-limit and current over-limit. [Results] Finally， the simulation based on the actual road network in Shanghai and the IEEE123-node distribution network shows that the proposed strategy can significantly enhance the resilience and operational safety of the system in fault scenarios. Compared with the mixed integer programming and particle swarm optimization methods， the method proposed in this paper demonstrates superior comprehensive performance in terms of load recovery rate， recovery speed， voltage stability and strategy security. [Conclusions] This paper verifies the effectiveness of hierarchical safe deep reinforcement learning in improving the resilience of ETCN. This method solves the problem of the separation of electrification-transportation targets through a two-stage architecture， achieving a balanced synergy among computing efficiency， load recovery rate and operational safety.

[Objective] To address limitations in spatio-temporal feature extraction and the integration of multi-scale features in wind power cluster prediction， this paper proposes a multi-scale fusion prediction method based on spatio-temporal graph spectral clustering （ST-SpecCluster） and a time-series 2D-reshaping Ttransformer （T2Dformer）. [Methods] A spatial skeleton is constructed utilizing the geographical coordinates of wind farms. A hybrid spatio-temporal graph is built to capture both static geographical proximity and dynamic operational correlations by dynamically calculating the Pearson correlation coefficients between wind farms using power data within sliding time windows. Deep spatio-temporal features are extracted via a spatio-temporal graph convolutional network （ST-GCN）， and an attention aggregation mechanism is combined with spectral clustering to partition the wind farms into highly correlated sub-clusters. For each sub-cluster， the power and numerical weather prediction （NWP） sequences are decomposed into trend， periodic， and residual components using the seasonal-trend decomposition procedure based on loess （STL）. A novel prediction model， the T2Dformer， is then designed to jointly model multi-scale features through the collaboration of fast Fourier transform （FFT） period extraction， time-series 2D reshaping， Inception convolution， and attention mechanisms. Each component of every cluster is predicted separately， followed by aggregation and reconstruction. [Results] The proposed method was applied to a wind farm cluster in Jilin Province， China. Compared with state-of-the-art prediction methods， the proposed approach reduced the normalized root mean square error by 1.89%， reduced the normalized mean absolute error by 2.87%， and increased the coefficient of determination （R²） by 10.87%. [Conclusions] This paper provides an effective solution for high-precision wind power prediction， with practical significance for enhancing the dispatching capability of power systems and the consumption of renewable energy.