What's new

Aiming to address the low efficiency of current deep learning algorithms for segmenting citrus in complex environments, this paper proposes a study on citrus segmentation algorithms based on a multi-scale attention mechanism. The DeepLab V3+ network model was utilized as the primary framework and enhanced to suit the characteristics of the citrus dataset. In this paper, we will introduce a more sophisticated multi-scale attention mechanism to enhance the neural network’s capacity to perceive information at different scales, thus improving the model’s performance in handling complex scenes and multi-scale objects. The DeepLab V3+ network addresses the challenges of low segmentation accuracy and inadequate refinement of segmentation edges when segmenting citrus in complex scenes, and the experimental results demonstrate that the improved algorithm in this paper achieves 96.8 % in the performance index of MioU and 98.4 % in the performance index of MPA, which improves the segmentation effectiveness to a significant degree.

One of the primary sources of noise and vibration in automobiles is gearboxes. Shafts, gears, and bearings are the main causes of noise and vibration in vehicle gearboxes. Various studies have reported that vibrations’ root cause is bearing excitation. Besides bearing fatal defects or extreme structure resonance amplification, gear mesh is the primary source of high-frequency vibration and noise, even in newly built units. Gear damage detection is frequently crucial in automotive gearboxes and vehicle safety. Furthermore, vibrations caused by shaft imbalances, shaft misalignments, and other factors can cause noise and vibrations in the drivetrain's transfer path. In addition, the vibration of an automobile gearbox is closely related to poor design, construction quality, and production accuracy. This paper reviewed previous research and methods on car gearboxes for conventional vehicles. It was obvious that frequency analysis and order analysis were commonly used in noise and vibration analysis on car gearboxes. Envelope analysis is usually used to analyze bearing faults. Finally, rolling-element bearing diagnostic techniques were also reviewed.

The lift leveling mechanism provided in this paper is controlled by the special rope fastening. By using the lift leveling mode controlled by the rope fastening, the leveling mode of the lifting system can be changed to the braking stop of the drive system after the platform is lowered and placed on the bracket. The problem of anchorage precision and the vibration caused by elastic expansion of the rope during platform loading and unloading are effectively solved by using the supporting function of the piers. In the process of studying this problem, we simplified a heavy-duty elevator system to form a simplified model that can be used for simulation research. The device provided in this article provides a new solution for the leveling of the heavy-duty lifting system, which not only ensures the leveling accuracy of the lifting platform during loading and unloading, but also effectively reduces the impact during leveling, which has a beneficial guiding effect on this type of equipment.

In order to ensure the good stability of the composite pipeline under fluid dynamic loads, finite element analysis method was applied to simulate and calculate the modal characteristics under dry and wet mode conditions. According to the working principle and structural characteristics of composite pipelines, the model was parameterized and simplified to improve computational efficiency and convergence. The modal testing platform is constructed, consisting of a combination of pipeline structure and pipeline support structure. The movement direction of the slider is limited by the guide rail used, allowing the horizontal section of the branch to move only along its own pipeline axis direction. The fluid structure coupling analysis technology was applied to obtain the vibration mode and spectral response results of the composite pipeline, which was verified through an experimental platform. The research results indicate that the natural frequency in wet mode was about 16 % lower than that in dry mode. In addition, it could be proven that the simulation scheme had good computational accuracy, with a maximum error of 9.2 %.

In order to ensure good stability and safety of the tower crane under long-term heavy load conditions, modal analysis and fatigue analysis were conducted on its key components, including standard section and frame section. The connection points of each truss structure were regarded as indivisible common nodes, and the model was simplified and parameterized, resulting in improved mesh quality and computational efficiency. Based on Pro/E, standard section and frame section models were established and imported into Workbench through intermediate data formats for simulation calculations in modal and static structural modules. By calculation, the natural frequency and modal shape of the standard section, as well as the lifespan, damage, and safety factor of the frame section, were obtained. It can be seen that the standard section has good structural stability and can maintain relative balance under excitation in different directions and degrees of freedom. If the safety factor of the web components is high and the service life is improved from a structural perspective, it can be considered to increase the thickness of the chord members.

Advanced aerospace structures are subjected to extremely harsh working environment, including of mechanical, acoustic, thermal, and aerodynamic loads. These combined loads can make the structures exhibit complex response, which has already attracted increasing concern in the design of advanced aerospace structures. To tackle the problem, a finite element model (FEM) by discretizing the theoretical differential equation is established to calculate the dynamic response of thin-walled structures under combined thermal and acoustic loads. The numerical analysis indicates that three types of nonlinear responses of the thin plate under combined thermal-acoustic loads are obtained, and the mechanism of snap-through is revealed through two ways: thermal buckling analysis and thermal modal analysis. which can be used to explain the nonlinear response characteristics.

In this article, the definition of the current density field at current collection was reduced only to the calculation of the thermal (resistive) action of the electric current. These procedures must include consideration of the bending stiffness of the contact wires to calculate the distribution of the pressing force on the contact wire between the current collecting plates. Therefore, it is necessary to improve the existing methods for calculating the dynamics of mechanical interaction of the current collector with the chain suspension. As a result of computer modeling in the Comsol Multiphysics software package, the volumetric power density of resistive heating from the intrinsic resistance of the contact wire was determined.

Adding of a multi-walled carbon nanotubes (MWCNTs) to epoxy resin has shown promising results in improving fracture toughness in bulk epoxy and carbon fiber-reinforced epoxy composites (CFRP). using a hand layup proceeding followed by the so called vacuum bagging process method, carbon fiber-reinforced polymer multi-wall carbon nanotubes (MWCNTs) was added to an epoxy resin with a weight percentage mixing of 1% wt., 1.25% wt., and 1.5 % wt. MWCNTs. Furthermore, the specimen underwent analysis via Fourier-Transform Infrared (FTIR) spectroscopy, and X-ray Diffraction (XRD) spectroscopy, the composites were subjected to a microscopic examination using a Scanning Electron Microscope (SEM). FTIR and XRD verified the folding and unfolding of the polymer, in addition, the mechanical properties including tensile strength, bending stress, and impact behavior were investigated as well as the hardness test. The obtained results showed a significant improvement of about (40 %) in tensile strength, (53 %) in bending stress at 1 % wt. MWCNTs, and (70 %) percentage increment in the strength of Impact at 1.25 % wt. MWCNTs. And the gained hardness was about 40.5 HV which were compared with a reference substance named Carbon Fiber (CF) without any addition of nano materials. Carbon nanotubes have demonstrated their potential to enhance the mechanical properties of fiber-reinforced polymers, so this investigative study employs comprehensive characterization techniques, and demonstrates significant improvements in mechanical properties for the modified polymeric composite materials supported with nano materials.