Acta Energiae Solaris Sinica•2026•DOI: 10.19912/j.0254-0096.tynxb.202608_9731
Fault diagnosis of wind turbine planetary gearboxes is severely constrained by the scarcity of high-quality fault data, as gearboxes operate predominantly in healthy states and automatic shutdowns prevent fault progression. This paper proposes a data-model jointly driven diagnosis method to address low diagnostic accuracy under limited fault samples. A high-fidelity lumped-parameter dynamic model of the planetary gearbox is constructed to generate pseudo-fault data, supplementing the training set. A domain-shared residual network feature extractor incorporating convolutional block attention modules extracts key physical features from both pseudo and measured data. Local maximum mean discrepancy aligns feature distributions at the fault-category level between pseudo and real fault data. A Kolmogorov-Arnold network module enhances the model's capacity to learn complex data relationships, enabling classification and identification of different fault types. Validation on a wind turbine planetary gearbox fault diagnosis test rig demonstrates that the proposed method achieves superior diagnostic performance under fault sample scarcity compared to classical methods. The framework offers an effective solution for known fault types, though identification of unknown and atypical faults remains a challenge for future work via open-set domain generalization.
SCIENCE CHINA Materials•2026•DOI: 10.1007/s40843-025-3615-1
Stretchable electronics are pivotal for bio-integrated devices, soft robotics, and wearables, yet their development is constrained by single-layer architectures that limit integration density and by mechanical mismatch between rigid components and soft substrates, which curtails service life. Here, we introduce a LEGO-like modular assembly strategy to construct multilayer three-dimensional (3D) stretchable electronics. Electronic components (ECs) and self-healing polyurethane (SPU) substrates patterned with liquid metal (LM) circuits serve as the modular blocks. This design simplifies fabrication and markedly enhances 3D integration density. The combination of LM circuits and self-healing elastic substrates enables devices to withstand diverse deformations and to autonomously heal after mechanical damage. Notably, the devices can undergo multiple recycling and reuse cycles without significant performance loss. This methodology offers a new paradigm for advanced flexible electronics, addressing critical bottlenecks in integration density, mechanical robustness, and sustainability.
Environmental Chemistry•2026•DOI: 10.7524/j.issn.0254-6108.2025032004
This study proposes an integrated source apportionment framework that synergistically integrates pollution source classification, atmospheric dispersion modeling, backward trajectory analysis, weighted trajectory clustering, and forward contribution estimation to accurately target peak reduction at localized air pollution hotspots. Applied at the County-Town Scale in Beijing, this method was employed to investigate pollution episodes at the Tongzhou Dongguan monitoring site. Source classification relied on a pollution fingerprint database and temporal concentration profiles, while local contributions were quantified through combined air quality modeling and monitoring data. Forward and backward trajectory analyses enabled the identification of potential source regions and key contributors. Results indicate that construction dust, road dust, and emissions from the catering industry were the dominant local sources, with construction and road dust contributing most prominently to PM2.5 concentrations. Furthermore, abnormal PM2.5 increases were closely linked to low boundary layer height, weak winds, and high humidity, emphasizing the role of meteorological conditions in pollution accumulation. The proposed framework proves effective in pinpointing local pollution sources and offers a scientific basis for targeted air quality management at finer spatial scales.
Journal of Fuel Chemistry and Technology•2026•DOI: 10.1016/S1872-5813(26)60670-6
Direct conversion of syngas to higher alcohols (C2+ alcohols) is critical for coal-based resource utilization and energy security. Here, a series of Na-modified CoFe/Al2O3 catalysts were synthesized via incipient wetness impregnation and evaluated for syngas-to-alcohol reactions. Multiple characterizations (XRD, N2 adsorption-desorption, H2-TPR, XPS, DRIFTS, in situ Raman, Mössbauer spectroscopy) elucidated synergistic effects of Na and Fe promoters. Na facilitated formation of Co-Fe alloy sites during reaction, while Fe modified electronic state of Co and promoted transformation of lattice oxygen to adsorbed oxygen, increasing surface oxygen vacancies. Synergistic interaction between alloy and carbide sites enhanced CO insertion into olefin intermediates, improving C2+ alcohol selectivity. Under 260 °C and 2 MPa, Co1Fe1Na1 catalyst (n(Co):n(Fe):n(Na)=1:1:1) achieved total alcohol selectivity of 45%, with C2+ alcohols comprising 94.1% of total alcohols. This study provides insights into rational design of Co-based catalysts for efficient syngas conversion to C2+ alcohols.
SCIENCE CHINA Materials•2026•DOI: 10.1007/s40843-025-4100-9
Shape memory droplet manipulation platforms have attracted significant attention due to their programmable droplet control capabilities. Current research primarily focuses on superhydrophobic surfaces and slippery lubricant-infused porous surfaces (SLIPS); however, these approaches suffer from vulnerable surface micro/nanostructures and loss of lubricant oils. Here, we report a shape memory quasi-liquid polydimethylsiloxane (PDMS) brush surface that overcomes these limitations. The surface is fabricated by introducing a SiO2 layer as a 'bridge' on a shape memory epoxy substrate, providing abundant functional groups for grafting PDMS brushes. By precisely controlling the SiO2 layer thickness and grafting conditions, the surface exhibits good shape memory properties and low adhesion to diverse liquids with varying surface tensions. Reversible anisotropic/isotropic droplet sliding control for both water and organic droplets is demonstrated through dynamic introduction/removal of groove structures, proving excellent droplet manipulation based on the combination of shape memory and low adhesion of PDMS brushes. Furthermore, the material can be applied as a functional coating on diverse substrates to impart anti-fouling and self-cleaning properties. This work introduces a nanoscale SiO2 layer as a 'bridge', offering a strategy to graft PDMS brushes onto polymer surfaces. Given the advantages of quasi-liquid PDMS brushes and programmable controllability of shape memory polymers, this work provides fresh ideas for developing droplet manipulation platforms.
SCIENCE CHINA Materials•2025•DOI: 10.1007/s40843-025-3286-2
Traditional thermocouples and infrared thermometers, while widely deployed, exhibit fundamental limitations in temperature range, accuracy, and environmental stability. Rare earth-doped fluorescent materials offer a non-contact alternative by exploiting temperature-dependent optical properties such as fluorescence intensity ratio (FIR) and lifetime. This review systematically analyzes the measurement principles, operational ranges, and application domains of fluorescent thermometry materials. Key material systems including Er3+/Yb3+ co-doped fluorides, oxides, and oxyfluorides are evaluated for their performance in biomedical and aerospace contexts. The review identifies selection rules for rare earth dopants and host matrices, with emphasis on FIR thermometry in visible and near-infrared regions. Critical parameters such as absolute sensitivity, relative sensitivity, and temperature resolution are compared across material platforms. The analysis reveals that core-shell structures and multi-ion co-doping strategies significantly enhance thermal sensitivity and photostability. For biomedical applications, the biological windows (1000–1700 nm) enable deep-tissue penetration, while for thermal barrier coatings, thermographic phosphors provide non-destructive turbine blade temperature mapping. The review concludes with recommendations for future development, including the need for standardized calibration protocols and scalable synthesis routes for industrial adoption.