中国电力科学研究院有限公司

Previous studies of lightning attraction have mainly focused on isolated tall structures, which can be approximated as point-like objects. Ultra-high-voltage (UHV) transmission lines form continuous elevated corridors, for which conventional point-source models are inadequate. Using 14 years of Lightning Location System (LLS) observations, this study examines changes in local lightning activity associated with the ±800 kV BJ UHV direct-current transmission line. A grid-based normalization method and a line-source lightning-attraction model were developed to compare lightning parameters before and after commissioning.Within 3 km of the line, cloud-to-ground (CG) flash density increased by 6.5%, with the increase decaying nonlinearly with horizontal distance. Return-stroke multiplicity and mean peak current increased by 6.8% and 7.3%, respectively, while the proportion of return strokes exceeding 40 kA increased by 11%. These current-related changes also weakened with distance from the line.Three spatial patterns of CG flash-density change were identified: rise–slow decay, rise–rapid decay, and rise–polarity reversal. Their differences were associated with aerosol concentration, topographic position index, large-scale terrain, and LLS location uncertainty. By extending lightning-attraction analysis from isolated point structures to a linear UHV corridor, this study introduces a line-source lightning-attraction model for transmission lines and links distinct lightning-density responses to environmental and observational factors. The results provide a basis for spatially differentiated lightning-protection design for UHV transmission systems.

With the accelerated transformation of the global energy structure, wind power, as a clean and renewable energy source, has attracted increasing attention. Its efficient utilization and sustainable development heavily rely on accurate power prediction. However, traditional forecasting methods face significant challenges due to factors such as turbine health conditions, environmental fluctuations, and data anomalies. To address these issues, this paper proposes a physics-informed bidirectional long short-term memory network (BiLSTM-PINN) based prediction framework. First, a method combining the Anisotropic Local Outlier Factor (A-LOF) with a Multilayer Perceptron (MLP), termed A-LOF-MLP, is employed to clean the raw data, effectively identifying and removing outliers and anomalous clustered data. Subsequently, based on the cleaned data, a Grey Relational Analysis-Long Short-Term Memory (GRA-LSTM) algorithm is used to assess the health status of the wind turbine. Finally, incorporating both health assessment results and physical constraints, the BiLSTM-PINN model is applied to achieve accurate wind power prediction. Experimental results demonstrate that the proposed A-LOF-MLP data cleaning method significantly improves data quality, and compared to existing approaches, the proposed framework exhibits superior performance in both prediction accuracy and robustness against disturbances.

High-precision interpretable forecasting of multi-energy loads is crucial for planning and optimizing Integrated Energy Systems (IES). Current methods have three key limitations: 1) reliance on closed-box models obscures feature interactions; 2) traditional modal decomposition methods ignore cross-modal correlations and fail to efficiently extract global cycle-trend characteristics; 3) limitations in sampling frequency prevent capturing fast dynamic characteristics of energy loads. To address these, we propose a dual resolution channel multi-period cross reconstruction parallel closed-form continuous-time network (DMP-PCFC) for multi-energy load forecasting. This method includes a multi-period cross-reconstruction network which equivalent to the signal sifter for decoupling original sequences and nonlinear mapping across periods, and models hidden states of energy loads using neuron state iterative differential equations. A parallel Closed-form continuous-time neural network (PCFC) based on chunked parallelization reduces inference time from $O(T)$ (change with input sequence length) to $O(P)$ (fixed cycle length). To overcome inherent sampling resolution limitations, a dual-resolution channel model with weight-sharing PCFC is used for simultaneous inference. Extensive experiments and continuous wavelet transform (CWT) on the IES dataset at Arizona State University's Tempe campus show that DMP-PCFC offers higher prediction accuracy and greater interpretability than existing methods, providing a new paradigm for downstream tasks based on other research methods, making it valuable for IES engineering applications. The code is publicly available at https://github.com/nuist-xf/DMP-PCFC

Spiking Transformers model token interactions primarily through spiking self-attention (SSA). However, binary query and key representations map continuous similarities to sparse and discrete relation responses, which may suppress weak relations and limit the propagation of local spatial context. To address this limitation, we introduce Spiking Local Interaction (SLI) and Adaptive Complementary Fusion (ACF). SLI establishes an attention-independent pathway for direct information exchange among neighboring spiking tokens using lightweight depthwise–pointwise transformations. ACF integrates SSA and SLI through layer-specific, channel-wise coefficients that adaptively balance their contributions at different network depths. The proposed design preserves the original attention formulation and can be incorporated into different Spiking Transformer architectures with modest parameter overhead. Experiments on ImageNet-1K, CIFAR-10, CIFAR-100, CIFAR10-DVS, and ADE20K show consistent improvements across image classification, event-based recognition, and semantic segmentation. In particular, QKFormer with SLI and ACF achieves 84.37% Top-1 accuracy on ImageNet-1K and 37.5% mIoU on ADE20K, where the segmentation model is trained without ImageNet pretraining. Ablation studies and qualitative analyses further indicate that SSA and SLI capture complementary interaction patterns and that learnable fusion consistently outperforms fixed weighting.

This letter primarily focuses on efficient angular estimation in low signal-to-noise ratio (SNR) conditions, as well as small-sample scenarios, aiming to facilitate the application of Integrated Sensing and Communication (ISAC) techniques. We propose a novel algorithm derived from multiple signal classification (MUSIC), called SS-MUSIC, which incorporates the sufficient statistic theorem. Simulation results demonstrate that SS-MUSIC achieves lower estimation errors in low-SNR conditions compared with existing MUSIC-based approaches, without increased complexity, and also exhibits superior performance in multi-target and small-sample scenarios.