Abstract:Hydrodynamic conditions play an essential role in algal growth and migration, but the mechanisms underlying algal responses to turbulence remain poorly understood. This paper systematically reviews the effects of turbulence on algal biomass accumulation and vertical migration, and further examines key factors affecting algal sensitivity to turbulent environments. First, turbulence regulates biomass by disrupting cellular processes, such as cell division and energy metabolism (photosynthesis and nutrient absorption). Second, turbulence alters vertical migration behavior by mediating algal buoyancy and mechanical stability. Finally, factors affecting algal turbulence sensitivity are analyzed in terms of cellular physiological structures and cell cycle phases. In view of current research gaps, future directions are proposed: deepening investigations into molecular regulatory mechanisms, establishing more comprehensive turbulence research frameworks, and improving coupled models linking algal physiology with turbulent physical structures to improve simulation accuracy. This review aims to provide theoretical support for understanding algal behavior under changing hydrodynamic conditions, developing bloom prevention and control strategies, and evaluating aquatic ecosystem services.

Abstract:Focusing on a hydropower-seawater desalination symbiosis system implemented in a coastal steel enterprise, this study addresses the challenges of using exhaust steam from a steam turbine as the heat source for low-temperature multi-effect desalination. In this highly coupled configuration between the steam turbine generator set and the desalination unit, coordinated safety control remains difficult and lacks mature technical solutions. To ensure equipment safety, this paper proposes a set of safety control strategies tailored to the symbiosis process, including start-up logic, equipment safety interlock protection under failure conditions, and mode-switching protection for the desalination unit under low steam turbine load. These strategies establish bidirectional interlock protection between the steam turbine generator set and the desalination unit. Application of the proposed control scheme in an actual coastal steel plant verifies its effectiveness. The strategy enables safe and orderly start-up and achieves bidirectional emergency shutdown when either subsystem fails, allowing desalination shutdown induced by turbine failure and vice versa. Concurrently, it accommodates large-scale load fluctuations in the plant’s gas supply, significantly enhancing operational safety.

Abstract:To develop a high-performance model for predicting the spontaneous combustion tendency of coal, on the basis of multiple gas indicators and industrial analysis parameters, four machine learning approaches (random forest, neural network, support vector machine, and Stacking ensemble) were used to predict spontaneous combustion temperature and natural ignition period, thereby evaluating coal spontaneous combustion risk. The findings indicate that the Stacking ensemble model exhibits superior generalization capability. Furthermore, feature importance analysis reveals that volatile matter and ethylene are the most influential predictors for natural ignition period and spontaneous combustion temperature, respectively. Model performance evaluation suggests that increasing data volume significantly enhances the predictive generalization of all four methods for spontaneous combustion temperature. However, expanding data alone yields only marginal improvement in predicting the natural ignition period. Enhancing feature representation is therefore necessary to further improve model performance.

Abstract:The corrosion evolution of oil and gas transmission pipelines is highly complicated, and sufficient data on influencing factors are often difficult to obtain in actual operation. Additionally, traditional empirical models produce significant errors in long-term predictions. To more comprehensively characterize the dynamic characteristics associated with memory effects and measurement randomness in pipeline corrosion, this paper proposes a non-Markov Wiener process prediction model considering both measurement errors and historical dependency. Model parameters are estimated and updated using maximum likelihood estimation and Bayesian inference. Based on weak convergence theory and the definition of first-passage failure time, an approximate analytical solution for the distribution of corrosion depth is derived, enabling predictive assessment of internal corrosion progression. Finally, monitoring data from the inner wall of a natural gas pipeline in the Chongqing Gas Mine are used to verify the effectiveness of the proposed method.

Abstract:To address the limited performance of existing chaotic systems in practical applications, this paper proposes a hyperchaotic system synchronization control method combining an interval type-2 fuzzy brain emotional learning controller (IT2FBELC) with a robust controller. The IT2FBELC approximates the unknown components of the hyperchaotic system, with its weights and parameters updated online via gradient descent to achieve synchronous tracking between the master and slave systems. The robust controller compensates for residual errors, driving the control output closer to the ideal value and further improving synchronization accuracy. Simulation results demonstrate that the proposed approach achieves high synchronization of hyperchaotic systems with superior tracking performance and computational efficiency compared to RBF neural networks, BP neural networks and conventional brain emotional learning models. Additionally, simulations for secure voice and image transmission confirm the method’s effectiveness and adaptability in confidential communication, providing theoretical support for practical applications of chaotic secure communication.

Abstract:To improve vehicle utilization and maximize resource efficiency in road freight transportation, this paper proposes a vehicle-cargo matching method based on view similarity, following case-based reasoning (CBR) principles. First, vehicle and cargo information is formally represented using a knowledge description system, enabling initial classification and matching through vehicle CR attributes and cargo N attributes. Subsequently, K-means clustering is performed on the vehicle dataset, and Mahalanobis distance is used to determine the cluster most similar to the cargo to be matched, thereby reducing the search space. An enhanced view-similarity calculation method is then introduced, where Euclidean distance is used to measure similarity between the target cargo and vehicles within the selected cluster. Experimental results show that the proposed method yields higher discrimination in matching results, with a maximum similarity of 0.848. Moreover, vehicle loading rates are significantly improved, with matching efficiency increased by about 76%. This method offers an effective approach for optimizing vehicle-cargo allocation in full-truck-load scenarios.

Abstract:Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a labeled source domain to an unlabeled target domain, playing a very important role in person re-identification. In real-world applications, video-based pedestrian data are often available, making it feasible to obtain single-camera-view labels in the target domain. However, existing UDA methods typically ignore this readily accessible information, thereby limiting performance improvements. To address this issue, we propose a knowledge-guided and fine-grained information enhancement framework for UDA person re-identification. A novel paradigm is introudced that leverages single-view labeled pedestrian samples in the target domain to fully exploit intra-domain information. Meanwhile, source-domain knowledge is used as guidance to assist the model to extract more discriminative target-domain pedestrian representations, effectively mitigating domain shift compared with conventional knowledge-transfer strategies. Furthermore, local pedestrian cues are integrated into global features to strengthen fine-grained feature expression. Experiments conducted on two publicly datasets fully demonstrate the effectiveness and superiority of the proposed method.

Abstract:To overcome the inherently narrow bandwidth of conventional microstrip antennas, this work proposes a wideband microstrip patch antenna using an L-shaped parasitic structure. The antenna consists of a main radiating patch and L-shaped parasitic elements, with an overall size of 0.83λ?×0.83λ?×0.083λ?, where λ? denotes the free-space wavelength at the 5 GHz center frequency. The main patch operates in the fundamental transverse magnetic (TM₁₀) mode, while multiple resonances are introduced via electromagnetic coupling with the parasitic elements, thereby significantly extending the bandwidth. A Rogers 5880 substrate with a relative permittivity of 2.2 and a loss tangent of 0.000 9 was used to reduce the quality factor. A 180° out-of-phase differential feeding scheme was applied to suppress radiation pattern distortion. Simulation results show that the proposed antenna achieves a voltage standing wave ratio (VSWR) below 2 from 3.6 GHz to 6.6 GHz, corresponding to a 60% fractional bandwidth—six times wider than that of traditional microstrip designs. The antenna exhibits a peak gain of 9.8 dBi at 6.2 GHz, with gain variation within 1.5 dB across the operating band. The main beam remains stable, with a pointing deviation within 5°, and cross-polarization levels remain below -15 dB over 3.6 GHz to 6.3 GHz, reaching as low as -38 dB at the center frequency. This single-layer configuration achieves both broadband and high-gain performance, demonstrating strong potential for broadband wireless applications, including 5G communication and Wi-Fi systems.

Abstract:To improve the efficiency of organizing and retrieving safety hazard information and to support more complex information processing tasks, effective technical methods for automatic text classification and type analysis are required. Support Vector Machine (SVM) can automatically classify unstructured text. However, their underlying principle focuses on identifying optimal classification boundaries within the training set and does not facilitate the extraction of representative features for each text category. To address this limitation, a normalized entropy model is proposed to search for typical category features, thereby improving the traditional term frequency-inverse document frequency (TF-IDF) based feature recognition method. Using 2 534 law enforcement inspection records from a government emergency management bureau as a case study, SVM was used for automatic text classification and achieved an accuracy of up to 97%. Meanwhile, the normalized entropy model was used to extract representative features for each category, providing decision support for formulating targeted rectification strategies in hazard investigation. Experimental results show that the combined use of SVM and the normalized entropy model effectively addresses both text classification and category feature recognition tasks.

Abstract:Lightweight design of orthodontic dental casts is critical for reducing the cost of invisible orthodontic treatments. In this study, deformation characteristics of actual dental casts were analyzed using the finite element method, and three structural infill strategies (hollow, honeycomb, and cube) were applied for weight reduction. Film lamination was simplified as a normal load applied to the outer surface, and the overall deformation response was used to evaluate lightweight performance. Finally, a newly fabricated dental cast was used to verify the simulation results. The findings indicate that the honeycomb infill structure provides the optimal lightweight effect for orthodontic dental casts, followed by cubic and hollow structures.