Recognition of flanges on LNG (“Liquefied Natural Gas”) vessels and distance determination are key prerequisites for automated docking of an unloading arm, and in complex environments, flange detection faces challenges such as low accuracy and large distance measurement errors. Therefore, an improved lightweight high-precision approach based on YOLOv8s that integrates the C2f_Ghost module, a CBAM, and Concat_BiFPN was proposed. In addition, a monocular ranging algorithm based on pixel width and coordinate transformation was introduced to estimate the 3D coordinates of a flange. Specifically, the original Bottleneck in the C2f module was replaced by the Ghost module, which combines dynamic convolution and dynamic depth-separable convolution to enhance feature representation while reducing model complexity. A CBAM was introduced in the middle layer of the backbone module to improve the model’s focus on key features with minimal parameter increases, and Concat_BiFPN was used in the neck to facilitate cross-scale feature fusion. To ensure the reproducibility of the experiments, this study primarily employed a fixed random seed (0), and experimental validation analysis was conducted on a flange dataset. The results show that for the improved model, mAP@0.5 reached 97.5% and mAP@0.5:0.95 reached 82.3% with a parameter size of 9.34 M, representing improvements in accuracy of 0.6% and 13.4%, respectively, and a reduction in the parameter size of 16.2% compared with the original YOLOv8s. The average ranging errors of the X-axis, Y-axis, and Z-axis were 2.43%, 2.77%, and 0.71%, respectively. Therefore, the combination of the two algorithms significantly improves the detection and ranging accuracy of flanges in complex environments.

To address the industry pain points of domestic traditional fish meal processing equipment, such as low protein retention, low drying efficiency, and poor operational reliability, this study focuses on high-moisture, heat-sensitive cod meal as the test material to investigate the structural improvement and synergistic optimization of process parameters for vacuum low-temperature fish meal dryers. The conventional uniform-pitch heating coil was optimized into a three-section differentiated structure, with a wear-resistant protective structure additionally incorporated to fundamentally resolve issues including insufficient heat transfer at the feed end, coking at the discharge end, and coil wear-induced leakage. Verification via COMSOL Multiphysics simulation revealed that the axial temperature gradient of the optimized equipment decreased from 8.6 °C/m to 6.2 °C/m, while the thermal fatigue life of the coil was extended from 2–3 years to over 10 years. A three-factor, three-level response surface methodology (RSM) was employed to design the experiments, with the heating temperature, vacuum degree, and drying time as independent variables and the fish meal protein content as the response variable. A total of 17 experimental runs were constructed, including 12 factorial points and 5 central points; each run was replicated three times in parallel, and data were reported as mean values. Analysis of variance (ANOVA) demonstrated that the regression model was highly statistically significant (p < 0.0001), with a coefficient of variation (CV) of 0.2464% and a coefficient of determination (R²) of 0.9944, indicating excellent fitting accuracy. The determined optimal process parameters were as follows: a drying temperature of 65 °C, vacuum degree of 0.08 MPa, and drying time of 75 min. Compared with the traditional process, the optimized process shortened the drying cycle by 37.5%, reduced unit energy consumption by 29.2%, and increased the fish meal protein content by 6.6%. This research provides a reliable technical solution for the localized processing of high-end fish meal.

High friction and drag are among the challenging subjects for constructing water-based drilling fluids available in horizontal drilling. Lubricants play a major role in mitigating friction of water-based drilling fluids, and thus, developing new lubricants is necessary for efficient horizontal drilling. In this work, a generation 1.5 (1.5G) hyperbranched polyamide-amine P(EDA-MA-OA), which serves as a candidate for a traditional lubricant with linear conformation, was newly synthesized via a divergent approach. A set of physicochemical characterizations was carried out on P(EDA-MA-OA) to confirm its effective synthesis. The results indicated that P(EDA-MA-OA) has a nanoparticulate morphology with a size of approximately 100 nm. Its molecular structure shows strong thermal stability, with initial thermal decomposition occurring at 146 °C. The water-based drilling fluid formulated with P(EDA-MA-OA) as the lubricant exhibits effective comprehensive properties and, in particular, the lubrication coefficient was 0.067, comparable to that of the oil-based drilling fluid, indicating enhanced lubricity by the incorporation of the hyperbranched polymer. The results of molecular simulations show that P(EDA-MA-OA) possesses a unique “basket-like” architecture, with C18 long chains enveloping the central active segments, namely the carbonyl (-C=O) and amide (-CO(NH₂)) groups. When interacting with montmorillonite (MMT) particulates, the active groups can interact with MMT, allowing the eight C18 branched terminal chains to form a “molecular brush” with a normal orientation toward the MMT interface, which can serve as a hydrophobic lubricating film to improve lubricity. A lubrication model was finally proposed to rationalize the enhanced lubricity from the hyperbranched polymers in the water-based drilling fluid.

To autonomously dock LNG unloading arms under adverse sea states, this study formulates a dynamic docking process as a trajectory forecasting task. By integrating visual-perception-based spatial localization with trajectory acquisition and forecasting, a comprehensive operational pipeline is established. To predict the dynamic trajectory of the vessel flange, an improved KSE-GRU model is proposed. By extracting implicit kinematic features, the model effectively enhances trajectory characterization under extreme sea states, thereby significantly improving forecasting accuracy and worst-case error constraints. To ensure the operational feasibility of autonomous docking, a robust control strategy is introduced to complement the trajectory predictions. The experimental results demonstrate that the proposed model outperforms traditional time-series forecasting models across all evaluation metrics. Compared with the baseline neural network models, the Mean-3D error is reduced by 19.14%, and the Max-3D error is capped at 348.77 mm, representing an 8.8% improvement over the baseline. Furthermore, the model demonstrates clear advantages in maintaining trajectory consistency and forecasting reliability. In summary, in this study, a robust trajectory forecasting model is developed for vessel target flanges integrated with a comprehensive control framework, thereby offering a practical approach to autonomous docking under dynamic oceanic conditions.