Search

In addressing the challenge of ships being prone to occlusion in multi‐target situations during ship target detection, leading to missed and false detections, this paper proposes an enhanced ship detection algorithm for YOLOv7‐tiny. The proposed method incorporates several key modifications. Firstly, it introduces the Convolutional Block Attention Module in the Backbone section of the original model, emphasizing position information while attending to channel features to enhance the network's ability to extract crucial information. Secondly, it replaces standard convolution with GSConv convolution in the Neck section, preserving detailed information and reducing computational load. Subsequently, the lightweight operator Content‐Aware ReAssembly of Features is employed to replace the original nearest‐neighbour interpolation, mitigating the loss of feature information during the up‐sampling process. Finally, the localization loss function, SIOU Loss, is utilized to calculate loss, expedite training convergence, and enhance detection accuracy. The research results indicate that the precision of the improved model is 91.2%, mAP@0.5 is 94.5%, and the F1‐score is 90.7%. These values are 3.7%, 5.5%, and 4.2% higher than those of the original YOLOv7‐tiny model, respectively. The improved model effectively enhances detection accuracy. Additionally, the improved model achieves an FPS of 145.4, meeting real‐time requirements. [ABSTRACT FROM AUTHOR]

Abstract Process energy consumption and environmental impact have been considered as important performance indicators for the sustainable electrical discharge machining (EDM) process. This paper presents a sustainable manufacturing technique known as magnetic field-assisted EDM (MF-EDM) to enhance the machine characteristics for the purpose of reducing the energy consumption and environmental hazards of the conventional EDM machining Ti6Al4V. Firstly, the principles of energy consumption, machining noise impact, and MF-EDM are respectively described. Thereafter, a set of experiments is carried out to investigate the effects of the main process parameters on the electrode wear rate (E W R), energy consumption (S E C), and environmental impacts (including carbon emission and machining noise) extensively, using the Taguchi method. The results indicate that the pulse on time (ranging from 100 to 200 μs) and magnetic field intensity (ranging from 0.05 to 0.10 T) are the two most significant factors affecting the MF-EDM sustainable manufacturing performance. Furthermore, the optimal process parameters were obtained by optimizing the MF-EDM process economically and environmentally using the modified non-dominated neighbor immune algorithm (M-NNIA) method. Compared to the minimum outputs of the experimental results, the optimal solutions of the E W R , S E C , and machining noise were significantly decreased, by 18.30%, 61.43%, and 20.95%, respectively. Therefore, it can be concluded from the above research that the proposed hybrid MF-EDM technique offers significant advantages and potential for applications in the sustainable manufacturing field. Graphical abstract Image 1 Highlights • A sustainable technique, magnetic field assisted EDM, is proposed to reduce energy consumption and environmental hazards. • Electrode wear rate, energy consumption, machining noise of MF-EDM processing Ti6Al4V were completely investigated. • Optimal parameter combinations were obtained through optimizing MF-EDM process economically and environmentally by M-NNIA. [ABSTRACT FROM AUTHOR]

SiC particulate reinforced Al-based metal matrix composites with high volume fractions (HVF − SiC p / Al) is widely applied in many engineering fields due to combining outstanding properties of both metal and ceramic materials. It's a tough problem to process HVF − SiC p / Al with superior machining performance and minor environmental effect. Thus, in this paper, magnetic field assisted electrical discharge machining (MF-EDM) of HVF − SiC p / Al is proposed to improve the sustainable machining performance including productivity and aerosol emissions which is most harmful to operators' health. Based on Taguchi method, a set of experiments are settled to investigate materials removal rate (MRR), electrode wear rate (EWR), and aerosol emissions under MF-EDM process of 45% and 65% SiC p / Al. It shows that pulse current is the major factor, and magnetic field assisted technology significantly develops the surface integrity whereas it just slightly contributes to improve MRR, EWR, and aerosol emissions when the intensity of magnetic is in the range of 0.1T–0.2T. Additionally, an optimization algorithm combing Quantum-behaved Particle Swarm Optimization with Gaussian distributed local attractor (GAQPSO) and Back Propagation Neural Network (BPNN) is employed to provide optimal machining parameters for economic and environmental MF-EDM process of HVF − SiC p / Al. Compared to the average experimental data, for 45% SiC p / Al , the average optimal solutions of EWR, and aerosol emissions were decreased by about 10% and 15% whereas MRR was increased by about 6.6%; for 65% SiC p / Al , the average optimal solutions of EWR, and aerosol emissions were decreased by about 5.7% and 10% whereas MRR was increased about by 5.5%. It presents an effective work for EDM process of HVF − SiC p / Al with high sustainable performance. • Magnetic field assisted EDM of HVF-SiC p /Al is proposed to improve productivity and aerosol emissions. • Material removal rate, electrode wear rate, aerosol emissions, and surface integrity of MF-EDM were completely investigated. • The composition of emitted aerosols from MF-EDM of HVF-SiC p /Al is analyzed. • MF-EDM process of HVF-SiC p /Al is optimized economically and environmentally by BPNN-GAQPSO method. • Optimal parameter combinations for MF-EDM process of 45% and 65% SiC p /Al are recommended. [ABSTRACT FROM AUTHOR]

Ultra-thin curved (2.5D, 3D) glass is widely used in the most recently developed smartphones, and the market demand for it is steadily increasing. The high-precision ultra-thin glass molding process (UTGMP) is very important in the fabrication of high-quality 2.5D/3D ultra-thin glass for smartphones; however, the UTGMP, with its high energy consumption and low pass rate, is not considered to be environment-friendly. Therefore, in this study, the sustainability of the molding process for 3D ultra-thin glass was comprehensively investigated. First, the ultra-thin 3D glass molding process is described, including a discussion of the glass molding apparatus, and the principle of glass thermos-formation is explained. Secondly, a thermo-mechanical coupling model is proposed to study the effects of loading heat and pressure on the distributions of temperature and internal stress of ultra-thin glass. The results demonstrate that significant amounts of stress were concentrated at the edges, and in holes and slots. A series of experiments were conducted to reveal the respective effect of various UTGMP parameters on the energy consumption, pass rate, and carbon emission level of the process. Consequently, the percentage of the main effect of loading temperature on energy consumption was found to be 23.65%, whereas that of the duration of applied pressure on the pass rate was determined to be 52.48%. The optimal process parameter combinations were determined by analyzing the above-mentioned research results and subsequently implementing them in a hybrid intelligent optimization algorithm that utilizes quantum-behaved particle swarm optimization (QPSO) and a back-propagation neural network (BPNN). The algorithm revealed that energy consumption and carbon emissions can be respectively decreased by 3.39% and 3.45% at a relatively high pass rate (i.e., over 70%). Eventually, the suggested combinations for the UTGMP will be able to achieve at least 60%–76% of the pass rate, as well as consume no more than 0.756 kW ⋅ h/pcs to meet sustainable manufacturing requirements. Thus, this study contributes to the effective application of a high-precision glass molding process for the sustainable manufacturing of ultra-thin curved glass. Image 1 • A thermos-mechanical coupling model of ultra-thin glass molding process (UTGMP) is proposed. • Distribution of temperature and inner stress of ultra-thin glass are simulated and analyzed. • Energy consumption, pass rate and carbon emission of UTGMP are completely investigated. • UTGMP process is optimized economically and environmentally by QPSO-BPNN. • Optimal parameter combinations of UTGMP are recommended. [ABSTRACT FROM AUTHOR]