Nuclear and Radiation Safety Center

Although Adaptive Traffic Signal Control (ATSC) can alleviate congestion issues to some extent in traditional signal control systems, it still faces challenges in dealing with complex and dynamic traffic environments, such as difficulties in agent coordination, high computational complexity, and unstable optimization results. To address these challenges, this paper proposes a multi-agent deep reinforcement learning algorithm based on SENet, called SE-A3C. The SE-A3C algorithm enhances the feature extraction capability and adaptability of the neural network by introducing the Squeeze-and-Excitation (SE) module from SENet. This allows the model to focus more precisely on high-information features and capture interdependencies between different channels, thereby improving the model’s discriminative ability and decision-making performance. Additionally, the algorithm incorporates Nash equilibrium concepts to maintain a relative balance among agents during coordinated control, avoiding suboptimal competition between agents and significantly improving system stability and efficiency. Experimental results show that, compared to traditional A3C, DQN, and Ape-X algorithms, the SE-A3C algorithm significantly improves the efficiency of traffic signal control and the overall throughput of traffic flow in complex traffic scenarios.

Cross-seasonal and cross-regional operations make it inevitable for low-temperature cycling of lithium–ion batteries, which accelerates battery aging and induces large inconsistency between batteries in the battery pack. This causes slight overcharging. However, the influence of long low-temperature cycling on the following slight overcharging aging and aging mechanism under multi aging path is not studied clearly. This affects the function of the battery management system (BMS), including state of health (SOH) prediction, state of charge estimation, etc. This work takes 18,650-type batteries as the study objects. Battery aging at low temperature (−10 °C) and slight overcharging (4.4 V) aging after low-temperature cycling are studied in this work. Hybrid pulse power characteristic, incremental capacity analysis, scanning electron microscope, and X-ray diffraction are used to reveal the aging mechanisms. The results indicate that a negative electrode degradation affects the cycle life of batteries more compared to a positive electrode, and the primary aging mechanisms are “dead lithium” and electrolyte decomposition. Compared to low-temperature cycling, slight overcharging is the lower stress factor. Cycling at low stress factor suppresses aging of battery cycled at high stress factor. When the SOH of battery is near 90%, lithium plating growing at low temperature is consumed after slight overcharging cycling. The generated products suppress further lithium plating. When the SOH is near 80%, although lithium plating is consumed, it also grows continuously. Slight overcharging causes more transition metal dissolution and graphite exfoliation. When SOH is near 90%, thermal management strategies should operate to control operation temperature of battery to avoid further low-temperature cycling. The results in this work are important to battery design and battery management system development.

The blade is a core component of the gas turbine, and blade fouling is characterized by highly concealed failure modes in the early stages and significant destructive potential in later stages. To address the lack of intelligence in early warning systems for compressor fouling, this study proposes a data-driven approach combining a digital-twin-based dynamic simulation model with the Weibull Proportional Hazards Model (WPHM) algorithm to enable reliable fault early warning. A modular design methodology was first adopted to construct a digital gas turbine model of the gas–gas combined power system on a dynamic simulation platform. High-fidelity fault simulation data were then generated to represent both healthy and faulty operating conditions. Through data governance and uncertainty quantification, key parameters influencing compressor fouling were identified. The Pearson correlation coefficient was applied to screen the most sensitive indicators, ensuring effective input selection for the prognostic model. Using historical health data from the simulation platform, the WPHM algorithm was trained to learn degradation patterns and establish a baseline failure risk model. This trained WPHM was then deployed to monitor real-time performance trends and provide early warnings for compressor blade fouling. Validation results from multi-unit simulations show that the proposed method achieves a fault warning rate of 95.0%, demonstrating its effectiveness and readiness to meet practical engineering requirements.

The parameter optimization of marine hydrodynamic models currently relies predominantly on expert empirical knowledge, but the quantitative indicators and weighting mechanisms for rapid calibration remain unclear due to inherent model uncertainties and complexities. This study addresses these challenges through expert questionnaires that collect fuzzy evaluations of calibration criteria, developing an integrated methodology combining the theory of axiomatic fuzzy set (AFS) with principal component analysis (PCA). Numerical case studies quantify calibration indicator weights and assess critical parameter impacts, revealing that bathymetry and roughness coefficients predominantly govern simulation accuracy. Elevated roughness conditions demonstrate two regimes: (1) at 1–2 × baseline roughness, strong positive correlations (with a coefficient of determination R2 increased by up to 0.568 compared to baseline) confirm effective model-data matching for tidal levels/currents; (2) beyond 2 × baseline roughness, progressive correlation decay accompanies increasing coefficients, indicating amplified simulation–measurement discrepancies. Notably, under reduced roughness conditions, high accuracy persists during spring/mid-tide phases but significantly diminishes during neap tides, demonstrating enhanced roughness sensitivity in low-tidal energy regimes.

This paper presents a dual-structured convolutional neural network (CNN) for image classification, which integrates two parallel branches: CNN-A with spatial attention and CNN-B with channel attention. The spatial attention module in CNN-A dynamically emphasizes discriminative regions by aggregating channel-wise information, while the channel attention mechanism in CNN-B adaptively recalibrates feature channel importance. The extracted features from both branches are fused through concatenation, enhancing the model’s representational capacity by capturing complementary spatial and channel-wise dependencies. Extensive experiments on a 12-class image dataset demonstrate the superiority of the proposed model over state-of-the-art methods, achieving 98.06% accuracy, 96.00% precision, and 98.01% F1-score. Despite a marginally longer training time, the model exhibits robust convergence and generalization, as evidenced by stable loss curves and high per-class recognition rates (>90%). The results validate the efficacy of dual attention mechanisms in improving feature discrimination for complex image classification tasks.