Guangdong Polytechnic Normal University

Online education in creative disciplines often misaligns with industry innovation. This study proposes an artificial intelligence (AI)-generated content-driven closed-loop model that synchronizes classroom creativity with industry demands via a unified digital platform. Codeveloped with three universities and 10 creative enterprises, the framework integrates a dual-mentor mechanism and real-time semantic tuning to establish a dynamic industry–academia ecosystem. Results show a 71% reduction in prototype iteration time, enhanced creative translation, and improved collaboration. Key scaling factors include algorithm compatibility, assessment fairness, and resource sustainability. An “open-incremental-collaborative” optimization strategy and a concrete AI governance framework covering transparency, bias mitigation, privacy, and intellectual property is proposed. This provides an empirically validated blueprint for AI-generated content-enhanced data-driven education.

Abstract Remaining Useful Life (RUL) prediction is a critical technology in prognostics and health management. Rotating machinery plays an indispensable role in industrial processes, and accurate RUL prediction for such equipment is essential to ensure production continuity, optimize maintenance strategies and reduce operational costs. In recent years, deep learning has emerged as a powerful tool for the analysis of multi-source monitoring data derived from rotating machinery, owing to its exceptional proficiency in feature extraction and non-linear modelling. This approach can effectively excavate the deeper-level information within the data, thereby improving prediction accuracy. Consequently, deep learning demonstrates significant application value and broad development prospects in RUL prediction for rotating machinery. This paper provides a comprehensive review of the application of deep learning techniques in RUL prediction for rotating machinery. Firstly, it introduces a unified framework for RUL prediction based on deep learning methodologies. The article then elaborates on the theoretical foundations of several key learning methods, including autoencoders, convolutional neural networks, long short-term memory, gated recurrent units, as well as emerging models such as transformers, temporal convolutional networks and graph neural networks. Subsequently, the paper summarizes the applications of these deep learning methods in RUL prediction for rotating machinery, evaluating the specific problems each method addresses based on their distinct characteristics. Finally, it discusses the significant challenges that deep learning faces within this domain and outlines potential directions for future research.

ABSTRACT Predicting the modulus of short fiber‐reinforced composites is complicated by simultaneous variations in fiber length, orientation, and volume fraction during processing. This work refines the classical modified rule of mixtures (MROM) by introducing two concentration‐dependent weighting functions, and , which capture the evolution of effective fiber length, fiber‐fiber interactions, fiber orientation, and loading‐angle effects. A microstructural efficiency coefficient is further incorporated to describe reinforcement degradation from dilute to crowded fiber conditions. The model was validated using ethylene‐vinyl acetate/short carbon fiber composites with fiber volume fractions of 1.37%–16.80%. Young's modulus measured at loading angles of 0°, 45°, and 90° showed excellent agreement with predictions, with the refined formulation outperforming the classical MROM, particularly at higher loading angles. Overall, the refined framework provides a compact and physically grounded framework for accurately predicting stiffness and supports the design and optimization of short fiber‐reinforced polymer systems.

Carbon fiber-reinforced composite (CFRP) exhibits high specific strength and excellent fatigue resistance, demonstrating broad application prospects. Direct fabrication of microscale sensors on CFRP surfaces has significant implications for structural health monitoring and functional integration. However, fabricating sensors on CFRP remains challenging due to its inherently weak surface polarity, while existing printedcircuit technologies suffer from low resolution and complex procedures. In this work, a temperature sensor was fabricated directly on CFRP via pretreatment, aerosol jet printing (AJP), and sintering. High-precision Ag nanoparticles (AgNPs) traces were printed by optimizing the printing parameters (carrier/sheath gas flow rate and printing speed) through simulation and experimental analysis. Then, AgNPs circuits with good conductivity and adhesion performance were obtained by optimizing the sintering processes, such as sintering temperature and time. Finally, grid temperature sensors with a temperature coefficient of resistance (TCR) of 2.981 × 10–3/°C and a resistance change rate of 0.545 Ω·°C–1 were successfully fabricated. The sensor achieved precise temperature sensing over 25 ∼ 70 °C with a response time of less than 200 ms and retained good stability after 500 cyclic operation tests, demonstrating excellent temperature monitoring capability. This work demonstrates advanced sensor fabrication on CFRP, paving the way for its potential applications in microelectronics, sensing, and monitoring.