Background Al-based radiators are core components of offshore generator sets, and their anticorrosion performance directly affects the safe and stable operation and service life. Methods A novel chitosan (CTS) derivative (CTS-ZIM) was synthesized through an amidation reaction by using a zwitterionic imidazole compound as a modifying monomer. Significant findings CTS-ZIM shows a significantly enhanced inhibition performance, achieving an inhibition efficiency of 97.1% at a low concentration of 1000 mg L–1 for Al fin in a 1.0 M HCl solution. Weight loss method, electrochemical method, and physical characterizations demonstrate that the inhibitor effectively suppressed the corrosion process through adsorption on the steel surface. The introduction of the zwitterionic imidazole compound into CTS provides abundant ionic structures, unsaturated bonds (CO, CC, CN), Schiff base moieties, imidazole rings, and heteroatoms (N, O). These changes enhance its adsorption on metal surfaces, favoring the formation of a protective layer. Moreover, density functional theory calculations indicate that the introduction of zwitterionic imidazole can enhance the adsorption strength and electron-donating ability of the inhibitor. It is believed that this work can provide sufficient guidance for the precise design of efficient corrosion inhibitors.
Extreme climate risks and farmland structural constraints are intensifying agricultural vulnerability, making the improvement of climate-smart agriculture (CSA) essential for enhancing agricultural resilience and sustainability. High-standard farmland construction (HSFC) represents an important practice for advancing CSA by improving cultivated land quality and farmland infrastructure. However, existing studies have mainly examined HSFC’s effects on single outcomes, including yield, efficiency, or environmental performance, with insufficient attention to whether it can simultaneously advance multiple CSA objectives or generate synergies and trade-offs among them. To examine the multidimensional linkages between HSFC and CSA, this study proposes an integrated Pressure-State-Response-Objective (PSRO) analytical framework. Using 2003–2022 panel data for 122 counties in Hunan Province, we apply a continuous difference-in-differences (DID) model and mechanism tests to estimate HSFC’s policy effects and assess transmission through input optimization, technology upgrading, and cultivated-land habitat quality. Results show that (1) HSFC has a positive and significant policy effect on CSA, with estimated coefficients of about 0.011–0.012 across specifications; (2) HSFC positively influences CSA in pillars food security and climate adaptation, but negatively affects pillar carbon mitigation in the short term; (3) heterogeneous effects are evident across regions, with stronger impacts where regional development conditions and land-consolidation potential are more favorable; (4) CSA gains from HSFC arise from improved farmland production conditions, reduced per-output input pressure, and complementary technological and cultivated-land habitat improvements. This study links a land-consolidation mega-program with a multi-pillar CSA index under an integrated PSRO framework and offers guidance for coordinating food security, climate resilience, and carbon mitigation.
Ensemble methods have been the norm for anomaly detection. However, existing ensemble methods for anomaly detection have three main issues: (1) Lack of diversity, the base classifiers are of the same algorithm with different initializations, for example, decision trees or neural networks, achieving only sub-optimal results. (2) The predictive uncertainty is not well calibrated for reliable and explainable anomaly detection, where overconfident predictions for both correct and erroneous classifications are made, making the results unreliable and hard to explain. (3) Traditional ensemble methods (e.g., bagging) cannot effectively capture the distinct distributions of predictions from diverse base models. In this paper, we propose a Diversified and Heterogeneous Ensemble learning framework with Calibrated predictive Uncertainty estimation (DHE-CU). We utilize a multi-layer perceptron (MLP) as the meta-classifier to combine the confidence of diverse base models, thereby achieving more explainable anomaly detection. We devise a global diversity loss that considers a global measure of diversity for the selection and pruning of the base models. The MLP meta-classifier can capture the diverse and distinct distributions of predictions from base classifiers. We use a simple yet effective method to quantify the predictive uncertainty of the meta-classifier. We propose a weighted accuracy-uncertainty calibration loss for class-imbalanced data to effectively calibrate predictive uncertainties. Various datasets are used to perform experimental evaluation extensively. The proposed DHE-CU framework demonstrates strong ensemble learning ability, achieving an average improvement of 8.8
Energy harvesting technique is of significance to achieve the self-powered long-term real-time wind turbine blade monitoring. However, the power output of the existing devices at ultra-low rotational frequencies (< 1 Hz) of wind turbines is still not satisfying for high-power-consumption Internet of Things (IoT) applications. Thus, this work proposes a dynamic bistable electromagnetic energy harvester based on the magnetic suspension structure. With the dynamic bistability, the system resonant frequency can be reduced, and the average equivalent electromechanical coupling coefficient and magnet vibration velocity can be increased. Thus, the output power at ultra-low rotational frequencies can be significantly improved. The governing equations of the proposed harvester based on the rotating coordinates are derived to predict the system performance. The prototype is fabricated to validate the dynamic bistability effect and energy harvesting performance. The experiments show that the proposed dynamic bistability can significantly boost the maximum output power in the frequency region by up to 86%, and 1.2 times compared with the linear and nonlinear monostable energy harvesters. Compared to the current state-of-the-art devices with different energy transduction mechanisms in literature, the normalized power, power volume density, and power mass density are improved by up to 1.27 times, 60.85%, and 22.67%. Parametric studies are conducted to evaluate the influences of geometric design. Finally, the prototype is experimentally demonstrated to achieve the self-powered IoT application under the ultra-low-frequency (< 1 Hz) rotational excitations. This work introduces a high efficient rotational energy harvesting technique for self-powered wind turbine blade monitoring.
In recent years, the rapid evolution of large vision-language models (LVLMs) has driven a paradigm shift in mul timodal fake news detection (MFND), transforming it from traditional feature-engineering approaches to unified, end-to-end multimodal reasoning frameworks. Early methods primarily relied on shallow fusion techniques to capture correlations between text and images, but they struggled with high-level semantic understanding and complex cross-modal interactions. The emergence of LVLMs has fundamentally changed this landscape by en abling joint modeling of vision and language with powerful representation learning, thereby enhancing the ability to detect misinformation that leverages both textual narratives and visual content. Despite these advances, the field lacks a systematic survey that traces this transition and consolidates recent developments. To address this gap, this paper provides a comprehensive review of MFND through the lens of LVLMs. We first present a historical perspective, mapping the evolution from conventional multimodal detection pipelines to foundation model-driven paradigms. Next, we establish a structured taxonomy covering model architectures, datasets, and performance benchmarks. Furthermore, we analyze the remaining technical challenges, including interpretability, temporal reasoning, and domain generalization. Finally, we outline future research directions to guide the next stage of this paradigm shift. To the best of our knowledge, this is the first comprehensive survey to systematically docu ment and analyze the transformative role of LVLMs in combating multimodal fake news. The summary of existing methods mentioned is in our Github: https://github.com/Tan-YiLong/Overview-of-Fake-News-Detection.
