Substation failures can trigger large-scale power outages and cause substantial economic losses. However, the scarcity of fault data forces existing substation failure detection methods to rely on single-modality information, thereby limiting their perceptual capabilities, while aggregating data from multiple substations for model training raises significant privacy concerns. To address these challenges, we propose a remote substation fault diagnosis framework. Specifically, we introduce S-informer, a model that integrates heterogeneous data modalities through a specialized encoder and a cross-modal attention mechanism for effective feature fusion. A multi-scale temporal attention module is incorporated to enhance the model’s ability to capture temporal dynamics. To tackle the class imbalance problem in fault data, we design a dual-path decoder architecture that performs normal-mode reconstruction and fault-mode classification. In addition, we develop a similarity-aware federated aggregation strategy that leverages inter-substation similarity metrics and incorporates differential privacy mechanisms, enabling privacy-preserving model training across distributed sites. Experimental results on a real-world dataset demonstrate that the proposed method achieves an accuracy improvement of 2.79% over the multimodal Informer, while reducing training time by 39.5% compared to FedAvg.

Wood surface defect detection is plagued by challenges in multi-scale feature fusion, small object recognition, and other aspects. To address these issues, this paper proposes SGCDT-YOLO, an improved YOLOv11n algorithm integrated with the Selective Weighted Spatial Reduction (SWSR), Guided Expansion and Dimensional-Oriented Selective Fusion Enhancement Network (GEDOSFEN), Dual-Stage Gradient Receptive Field Aggregation (DSGRFA), and triplet attention modules. A dual-branch SWSR module is designed to enable content-aware downsampling, preserving discriminative information in critical defect regions while suppressing irrelevant areas. GEDOSFEN is proposed to build bidirectional diffusion paths via intermediate-scale features, facilitating effective interaction between deep semantic information and shallow texture details across detection scales. Inspired by dilated-wise residual and dilated re-param block, the DSGRFA module is designed to decompose receptive field expansion into regional feature construction and semantic filtering, extending the perceptual range with low computational overhead. Triplet attention is incorporated to establish cross-dimensional correlations among channel, height, and width, enhancing the unified perception of multi-morphology defects. Experimental results show that SGCDT-YOLO achieves 78.12% mAP@0.5 and 40.42% mAP@0.5:0.95, representing improvements of 13.41% and 5.43% over the baseline, respectively, with precision at 76.62% (+10.25%), recall at 68.52% (+6.76%), and F1-score at 72.34% (+8.36%).Meanwhile, the number of parameters is reduced by 10.47% compared with the baseline model. Comparative tests reveal a 12.22–16.02% mAP@0.5 improvement over mainstream algorithms. Visualization analyses of inference results, heatmaps, and feature maps confirm the algorithm’s high accuracy in defect localization and classification.

Superconducting resonators and qubits are limited by dielectric losses from surface oxides. Surface oxides are mitigated through various strategies such as the addition of a metal capping layer, surface passivation, and acid processing. In this study, we demonstrate the use of x-ray photoelectron spectroscopy as a rapid characterization tool to study the effectiveness of cap layers for niobium for further device fabrication. We non-destructively evaluate 17 capping layers to characterize their ability to prevent oxygen diffusion and the effects of standard fabrication processes—annealing, resist stripping, and acid cleaning. We downselect for resilient capping layers and test their microwave resonator performance.

The dynamics of low-energy proton collisions with hydrocarbon molecules are investigated using real-time time-dependent density functional theory. Through systematic variation of proton kinetic energy and impact site on the molecular surface, the resulting scattering, proton capture, and bond dissociation pathways are analyzed. The simulations reveal a strong dependence of reaction outcomes on both incident energy and collision geometry, with the interplay between electronic and nuclear degrees of freedom highlighted as governing molecular fragmentation and reaction mechanisms. These findings provide insight into the fundamental processes underlying proton–hydrocarbon interactions relevant to radiation chemistry, ion-beam processing, astrochemical environments, and Coulomb explosion.

The core size distribution is a critical parameter determining the performance of magnetic nanoparticles (MNPs) in biomedical applications. In this study, the magnetic interference of the hydrogel matrix on the core size estimation was investigated. It was identified that the matrix generates a diamagnetic response that superimposes on the superparamagnetic response of the MNPs. This superposition results in a distortion of the static magnetization curve, particularly in the high-field region. Consequently, the core size distribution estimated from the uncompensated magnetization deviates significantly from the true value. To address this issue, an improved method is proposed to accurately estimate the core size distribution by compensating for the matrix interference. Based on the identified superposition characteristic, a compensation model was established to subtract the diamagnetic contribution and recover the intrinsic magnetization of the MNPs. Experimental results demonstrate that the core size distribution estimated using the proposed compensation method is significantly more accurate than that obtained directly from composite magnetization, demonstrating the necessity of this strategy for the precise characterization of MNPs in hydrogel-based biomimetic environments.

Recent theoretical approaches have renewed attention to the idea that consciousness should be understood not as a by-product of physical processes but as a foundational aspect of reality itself. Panpsychism, foundational consciousness models, and related field-theoretic approaches share a common commitment to treating consciousness as cosmologic fundamental rather than an emergent one. However, many of these theories lack an explicit structural account of how concrete, private experience is generated from a non-dual foundation. This paper aims to address this gap by presenting an internal generative architecture that completes foundational consciousness approaches without rejecting their core premises. Instead of treating experience generation as a single continuous mapping, we introduce a three-layer structural model consisting of concordance potential, the Beral layer, and the phenomenal domain. The central contribution of this study lies in defining the Beral layer as a transparent mediating layer that generates neither meaning nor subjectivity, yet it enables experiential differentiation between the foundational and phenomenal levels. Furthermore, we argue that experience is generated through an irreversible three-stage structure consisting of (1) establishment of experience, (2) qualia formation, and (3) qualia acquisition. This distinction clarifies that qualitative properties of experience (qualia) can be generated prior to their attribution to a subject and that subjectivity or selfhood is not the cause of experience but emerges retroactively through the process of experiential acquisition. The proposed three-stage model provides a structural explanation for why artificial systems cannot acquire qualia, not in terms of insufficient capability or complexity but in terms of fundamental structural differences. In addition, by introducing the concept of Kamae as a pre-observational condition that precedes observation and meaning attribution, the framework explains how different experiences can arise from identical foundational states without reducing such variation to subjective arbitrariness. While this study supports approaches that regard consciousness as foundational, it is simultaneously critical of them. Rather than asserting that “consciousness is fundamental,” the focus is placed on the structural question of how experience is generated from a foundation. By introducing the Beral layer and the irreversible three-stage model, this paper offers a generative framework that describes experiential differentiation while preserving non-duality and proposes a coherent completion of generative approaches in consciousness studies.

Using triethoxy-1H,1H,2H,2H-tridecafluoro-N-octylsilane to modify hollow glass microspheres, fluorinated hollow glass with low surface energy and high reflectivity is obtained. Then, TiO2 is added, and epoxy resin E51 modified with silane coupling agent KH550 is used as the binder for uniform spraying. After drying, a coating with high reflectivity is obtained. The coating has a water contact angle of 154° and exhibits excellent cleaning ability for liquid and solid contaminants on the coating surface. At the same time, the coating has a solar reflectance of up to 92.5% and a high infrared emissivity, which can achieve a cooling effect of around 14 °C, exhibiting excellent radiative cooling performance. The coating maintains its self-cleaning and radiative cooling capabilities even after 120 wear cycles.

Self-Compacting Concrete (SCC) is widely used in modern construction due to its superior flowability, but its fracture behavior under complex loading conditions remains insufficiently understood, especially in the presence of pre-existing cracks. This study investigates the fracture performance of SCC under mixed-mode I-II loading using a numerical approach based on the Extended Finite Element Method (XFEM). Semicircular bending specimens with various initial crack angles were modeled, and results were validated against experimental data. The numerical predictions showed a maximum error of 8% in fracture loads compared to experimental values, and crack growth paths were accurately captured. The results indicate that increased crack angles lead to higher fracture resistance due to the contribution of shear-induced plasticity. These findings demonstrate that XFEM is a reliable tool for predicting fracture in SCC, offering practical insights for the design and assessment of cracked concrete structures.

To meet the requirement for the low disturbance of wind tunnel flow in supersonic boundary layer transition simulation tests, a low-disturbance supersonic research wind tunnel with a nozzle size of 0.3 × 0.15 m2 and test Mach number of 3.0 was built. It was then used to study key design technologies for increasing the size and Reynolds number of low-disturbance supersonic wind tunnel. The wind tunnel adopted a “blowdown + ejection” layout to achieve a wide pressure range of 50–450 kPa. A “sintered wire mesh + honeycomb + damping mesh” structure was adopted in the settling chamber with the flow straightening method to achieve velocity pulsations of less than 0.3% and pressure pulsations of less than 0.1% at the exit. A comprehensive design technique for the supersonic laminar nozzle was adopted, which included nozzle boundary layer suction and highly polished surface nozzle processing. At a total pressure of 450 kPa, a low-disturbance supersonic test area with an axial length of about 9 cm was realized. The airflow pressure pulsations in this area were less than 0.15%, and the Reynolds number based on the characteristic length reached 1.3 × 106.

Nonlinear dynamics plays a significant role in interdisciplinary fields spanning biology, engineering, mathematics, and physics. Under small-amplitude approximations, certain nonlinear systems can be effectively described by the linear Mathieu equation. Within the Lagrangian mechanics framework, here we systematically investigate nonlinear dynamics and parametric resonance phenomena in a collision-coupled double pendulum system. The complete dynamical model successfully captures the system’s nonlinear behavior under large angular displacements. Numerical simulations and experimental validations demonstrate parametric resonance within specific frequency ratio ranges, with resonance boundaries showing good agreement with theoretical predictions. The collision coupling mechanism enables unique energy transfer properties, while system nonlinearity ultimately leads to amplitude saturation. Notably, frequency modulation techniques achieve perturbation-free excitation of parametric resonance. We propose a novel example in nonlinear dynamics, with a methodological framework applicable to broader collision-coupled systems.