Addressing key technical challenges in surface defect detection for automotive alien wheels, such as the difficulty in identifying minute defects and insufficient detection accuracy. This paper proposes a method for detecting out-of-round defects based on array imaging and traditional visual inspection. First, a line-scan camera is used to capture images of the alien wheel surface. Through image fusion techniques and affine transformation algorithms, precise correction is Correcting the surface of irregularly shaped rotating surface, Effectively overcame the issue of positional deviation during workpiece rotation. Building upon this foundation, a defect detection algorithm based on an improved differential enhancement approach was designed, Through a dual strategy of background estimation and feature enhancement, the visibility of minute defects has been significantly improved. We further propose a dual-path spot analysis mechanism, implementing targeted detection on both the original image and the difference-enhanced image.

Visual Simultaneous Localization and Mapping (vSLAM) performance often degrades significantly due to interference from moving objects in dynamic environments. To enhance system robustness, this paper proposes a novel DYX-SLAM system. This paper deeply integrates dynamic perception and adaptive feature processing techniques. By employing instance segmentation networks to mask dynamic objects, introducing lightweight deep features to replace traditional handcrafted features, and designing a scene-content-aware adaptive threshold mechanism, it effectively addresses feature matching challenges in dynamic interference and low-texture environments. Experiments on public datasets such as TUM RGB-D and Bonn RGB-D demonstrate that while maintaining real-time performance, our system achieves absolute trajectory error (ATE) improvements exceeding 95% compared to ORB-SLAM3. These results demonstrate the system's superiority, robustness, and generalization capability in dynamic environments.

ObjectiveThe primary failure modes of aviation spiral bevel gears include tooth surface scuffing, wear, and scoring. Current evaluation methods for these failures neglect the influence of surface integrity parameters, making it difficult to accurately assess the correlation between the appearance of scuffing and the actual degree of damage. This study was conducted to establish a clear relation between surface integrity parameters and the extent of tooth surface damage.MethodsTaking a pair of aviation main reducer spiral bevel gears as the research object, the evolution of its surface integrity parameters—surface roughness, residual stress, and surface layer crystal structure—was systematically measured and analyzed pre- and post-scuffing. Specifically, the surface morphology skewness and kernel average misorientation maps at various depths were employed to quantitatively characterize the scuffing topography and the micro-plastic strain state. Based on the numerical changes in these integrity parameters, the actual damage state of the tooth flank was analyzed, leading to the proposal of a new concept and a precise evaluation method for a phenomenon termed “running-in scuffing”.ResultsFor the specific aviation spiral bevel gears studied, significant contact stress and high sliding velocity during test runs lead to tooth surface scuffing. The surface integrity parameters after scuffing are characterized by a decrease in surface roughness, a surface skewness approaching zero, an increase in residual compressive stress, and grain refinement. This type of scuffing is defined as “running-in scuffing”, and it is proposed that gears exhibiting these characteristics can continue in service and should not be classified as failed. This research provides a new perspective on the classification and evaluation of gear tooth scuffing and offers a reference for troubleshooting scuffing faults in aviation gears.

Inverse kinematics is one of the core issues in robot research, serving as the foundation for robot motion planning, trajectory generation and control. In this paper, a novel hybrid optimization algorithm is proposed to solve the inverse kinematics problem of robots effectively. It leverages the Quasi-Newton method to accelerate the convergence rate of the Electromagnetism-like algorithm (EM), and enhances its exploration capability for high-precision solutions. Calculating the decreasing direction of the objective function value and combining it with the random step size method for fast local search. Several experimental results indicate that the proposed method significantly improves solution accuracy while maintaining high search efficiency.

In order to improve the detection accuracy and stability of FPC connector defects, a machine vision detection method based on an improved Canny operator was proposed in this paper. Firstly, perform line detection through point fitting to calculate the tilt angle, and then perform rotation correction on FPC image. Next, Otsu algorithm was used to segment the filtered image, and an improved Canny operator was proposed to achieve edge detection, which can effectively suppress the effects of pseudo edges and edge fractures by utilizing sub-pixel information in the gradient direction. Based on the characteristics such as size, area, and position of the detection target, the defect target was segmented by subtracting it from the original image through opening operation. Moreover, a improved shape matching was used for detecting damaged defects, and the number of missing pin angles can be obtained simultaneously through position information. Finally, a corresponding FPC connector defect visual inspection device was developed. The experimental results show that the proposed method can achieve automatic detection and classification of common defects with a detection accuracy of over 95%.

Focusing on stress distribution at subsurface layer under aerospace service condition in roughness tooth flank meshing interface, a new loaded contact fatigue life forecasting model is developed by considering micro-scale surface effect for aerospace spiral bevel gears. Firstly, tooth flank modeling considering the actual manufacturing process is used for accurate tooth flank point determination having high and uniform grid density. Then, with application geometric approximation and operation, discrete convolution and fast Fourier transformation (DC-FFT) based conjugate gradient (CG) method is applied to determine time-varying load distribution. While at normal direction of each point from the high-density tooth flank discretization after accurate interpolation is added the roughness height from the actual micro-scale geometric topography measurement, a tooth flank reconstruction is performed to determine the micro- scale geometric topography. Then, elastic half-space loaded contact model and DC-FFT method are employed to compute subsurface stress distribution for roughness tooth flank. Von Mises stress is selected as design variable and introduced into Zaretsky model to establish the contact fatigue life forecasting model. Finally, a spiral bevel gear set in aerospace industrial application is exercised to verify the impact of subsurface stress on contact fatigue life.

The tooth surface geometry of harmonic gears directly affects the transmission accuracy and service life. Traditional design methods may cause tooth profile distortion when changing the radial deformation coefficient, which limits their application. This paper proposes a comprehensive tooth surface modification method that changes the radial deformation coefficient on the basis of traditional design methods. Firstly, the meshing trajectories and corresponding tooth profiles of gear teeth under different radial deformation coefficients are calculated and analyzed based on the rack approximation method. Secondly, a calculation method is proposed to eliminate the tooth profile distortion caused by changing the radial deformation coefficient, which not only expands the application of the rack approximation method but also eliminates interference during the meshing process. Subsequently, a comprehensive tooth surface modification method is proposed with the aim of increasing contact area and contact ratio, as well as reducing contact stress. Compared to traditional modification, it requires less material removal, which is beneficial for increasing the tooth strength. Furthermore, a finite element simulation model of a harmonic drive is established, and the tooth surface and contact performance of harmonic gears under three different radial deformation coefficients are designed and analyzed, verifying the effectiveness of the proposed tooth surface design method.

Directly modeling the inverse kinematics problem of robots using joint angles requires handling complex mapping relationship between joint parameters and poses. Unlike the traditional methods, a new node deployment method was proposed, and a distance-constrained graph model was established to re-characterize the problem. Meanwhile, the feasibility was demonstrated by establishing a precise conversion relationship between the parameters of the graph model and the DH parameters. Moreover, to enhance computational efficiency, the solution process was transformed from Euclidean space to Riemannian manifolds, and a gradient formula that can be used for optimizing Riemannian manifolds was derived in details. Finally, the performance of typical first-order Riemannian optimization methods were tested and compared in solving the robot inverse kinematics problem. The experimental results demonstrate the effectiveness of the proposed method.

In recent years, with the rapid advance in computing and deep learning technologies, more and more modern designs of neural network models are introduced. Numerous research has been conducted to explore the potential of these models on solving subsurface problems. Instead of investigating the performance of these developed models on subsurface applications, this paper aims to introduce a new concept that helps deep learning models identify the location of a source in a subsurface flow field. In this study, the concept of conservation field is introduced that is initially calculated by a simplified scheme for homogeneous domains and then the scheme is generalized using trained U-Net models based on a Gaussian kernel. Even with the simplified scheme, the source localization accuracy using conservation data is increased by more than 4 times compared to using pressure data directly. When using the Gaussian kernel-based approach with the presented postprocessing technique, the accuracy is further enhanced by two to three orders of magnitude. Even for datasets with different degrees of heterogeneity and uncertain number of source, the model trained with the homogeneous dataset of single-source still maintains a high accuracy. Furthermore, the presented probability learning approach can produce a series of possible source locations tied with probabilities, instead of only making single prediction. Analysis criteria are introduced to determine real source locations from all possible locations.

There are limited comparative studies on modeling fluid transport in fractured porous media. Hence, this paper systematically compares the steady-state creeping flow Stokes–Brinkman and Darcy–Darcy models for computational efficiency and accuracy. Sensitivity analyses were also conducted on the effect of fracture orientations, fracture sizes, mesh resolution, and fractures with Local Grid Refinement (LGR) under the FEniCS computational framework. Both models were validated numerically, and the accuracy of their solution is compared using the R-squared metric and L2 norm estimates. Key results showed that both models have similar pressure and velocity field solutions for a given fracture orientation. The computational time required for solving the Stokes–Brinkman models for a single fracture case was unusually lower than that of the Darcy–Darcy model when the pressure and velocity terms in the Darcy–Darcy model were solved simultaneously using two equations, contrary to where only one equation solves for the pressure and the velocity is obtained by projecting the gradient of pressure onto a vector space. The Stokes–Brinkman model is more sensitive to mesh resolution, and as a result, the Darcy–Darcy model tends to be more accurate than the Stokes–Brinkman model at low resolutions. Local Grid Refinement (LGR) can improve the Stokes–Brinkman model's accuracy at low mesh resolution. Furthermore, both models showed similar results when compared for complex fracture systems such as multiple fracture cases: interconnecting and isolated fractured porous media systems under low-velocity and steady-state creeping flow conditions. The FEniCS code in this paper is shared for future researchers to reproduce results or extend the research work.

Most numerical simulations for modeling acid reactive fluid transport and wormhole propagation during matrix acidizing, waterflooding, and CO2 sequestration in carbonate formations are computationally expensive, limiting real-time reservoir management and deep learning training datasets generation for inverse modeling research. Therefore, there is a need for less computationally expensive acid-reactive fluid flow models with adequate accuracy. This study developed and validated a simplified acid reactive-transport model by integrating a simplified Stokes–Brinkman model (as opposed to Darcy’s law), an averaged continuum model, and a pseudo-fracture model. Using FEniCS, the model effectively simulates acid-reactive fluid transport and wormhole propagation in carbonate rocks, achieving a high R-square value of about 0.97 based on a quantitative comparison of the breakthrough volume with other models. The simplified model can also simulate wormhole propagation for the reciprocal of the Damköhler number (1/Da) ranging from 0.001 to 1 with adequate accuracy. Sensitivity studies on the natural fracture parameters such as orientation, length, width, and density showed that higher fracture density, wider fracture aperture, longer fracture length, and orientation aligned with the direction of acid injection contribute to lower pore volume to breakthrough ratio but may not increase long-term acid stimulation efficiency. Also, the presence or absence of fractures in the matrix does not alter the dissolving patterns and optimum injection rate. This simple acid reactive-transport model can generate large training datasets for developing surrogate models in deep learning research. Finally, the FEniCS code in this paper is shared so future researchers can reproduce the results or extend the research work.

Effect of foreign direct investment (FDI) in services on carbon productivity is examined based on Chinese province-level panel data from 2007 to 2019. We control for income level, technical level and sectoral differences and find (i) in the full sample, service FDI has no significant effect on carbon productivity; (ii) the effect of income inequality: service FDI flowing into high-income regions benefits the environment, while flows to low-income regions do not support the pollutionhalo effect; (iii) the effect of technology inequality: service FDI in regions with high technology levels reduces the carbon emission, while the environmental effect of service FDI is insignificant in low-technology regions; (iv) industrial heterogeneity: FDI inflows to different sectors show different sensitivities to income level and have different impacts on carbon productivity. The result has a strong policy implication which will be helpful for the local governments to make policies with the aim to maximize the benefit from the pollution-halo effect of service FDI.

The Vickers hardness measurement method is one of the most commonly used methods of measuring the hardness of materials.Traditional manual measurement is not only labor-intensive but also influenced by the material surface, making it difficult to guarantee measurement accuracy. In contrast, automatic Vickers hardness measurement using image processing technology can significantly reduce labor intensity and is highly robust in complex measurement environments. The imaging, pre-processing, and acquisition of the indentation vertex position of the Vickers indentation image are key aspects of visual measurement. This paper presents the main research advances in the relevant content and analyses the characteristics and shortcomings of existing methods. It also discusses the development trends of research in this technical field.

The automatic measurement algorithm for Vickers hardness indentation has been widely applied. Among these, the neural network-based method has received attention for its excellent segmentation performance. However, high storage space and computation requirements hinder its promotion on edge computing devices. To address this issue, this study proposes two lightweight Vickers indentation segmentation networks: VSNLite4M and VSNLite1M. Compared with previous methods, the proposed networks achieve a reduction of 35.2x in terms of computational cost with up to 38x fewer parameters, while maintaining the same level of segmentation accuracy.

vSALM is a system that enables precise positioning of a mobile robot in an unknown environment and generates corresponding 3D maps. However, the feature matching process of the visual odometry in the SLAM system is easily disturbed by dynamic objects in real-world scenarios, such as moving pedestri-ans and vehicles, which will reduce the accuracy of positioning and mapping. To address this issue., a visual SLAM system with dyna-mic object removal., called as Dynamic Object Removal SLAM (DOR-SLAM)., is proposed based on the ORB-SLAM2 system. Firstly., a dynamic probability calculation formula of pixels based on optical flow and semantic information has been proposed to determine the motion state of object. Then., a process of instance segmentation is used to segment the recognized dynamic objects in the image frames. Next., these dynamic objects will be removed by using a flow-guided video inpainting network., while inpainting the background. To improve system efficiency., a simplified bidirecti-onal optical flow calculation method is proposed and applied to the video inpainting network. The effectiveness of the proposed meth-od is evaluated with dynamic RGB-D sequences from the TUM dataset. The results of the experiments indicate that the proposed DOR-SLAM can improve the accuracy of localization in dynamic environments while maintaining high computational efficiency.

高校在培养目标、毕业要求和课程体系的制定过程中,应围绕如何坚持立德树人、引导学生树立社会主义核心价值观这一核心问题展开.该文根据"树立和践行社会主义核心价值观"的目标和任务要求,探讨了包含社会主义核心价值观的毕业要求指标点有效分解和表达方法,形成了构建社会主义核心价值观培养体系的总体设计思路.相关研究成果对于在工程教育专业认证过程中有效落实社会主义核心价值观的培养具有积极意义.

In this paper, the model and its curriculum learning method of garbage hierarchical classification and corresponding operation mode decision in home environment are proposed from the perspective of cleaning robot. In order to realize the hierarchical learning of garbage attribute concept, this paper designs a learning model with iterative feedback network as the backbone network. In the early stage of iteration, the model focuses on learning the state of garbage, in the middle stage, it focuses on the appearance attributes of garbage, and the specific categories of garbage in the later stage. At the same time, the attention module is introduced to achieve different levels of feature expression learning, which further improves the performance of the model. The evaluation was conducted on the collected garbage data set and the public CIFAR-100 and Stanford Cars data sets, which verified the effectiveness and wide applicability of the proposed method.

Using analogue reservoirs for comparison and benchmarking is a comprehensive method for ensuring accurate measurement and gaining a better understanding of potential oil recovery enhancement prospects. The approach of leveraging information from analogue reservoirs is extended to CO2 EOR project management in this study. We present a novel application of machine learning clustering algorithms to the rapid identification of analogues for new projects without having to sift through massive amounts of data. We use machine learning clustering methods to group successfully executed miscible CO2 flooding projects into clusters of projects with similar fluid/reservoir characteristics and to identify analogues for new target projects. Porosity, permeability, oil gravity and viscosity, reservoir pressure and temperature, minimum miscibility pressure (MMP), and depth were all input parameters. Data from nearly 200 miscible CO2 EOR projects around the world were clustered using the Agglomerative Hierarchical Clustering Algorithm (HCA), K-Means, and K-Median techniques. To reduce information redundancy in high-dimensional data, Principal Component Analysis was used as a pre-processing step. Three evaluation indices (the Davies Bouldin, Calinski Harabasz, and Silhouette Coefficient Scores) were used to compare the efficiency of the clustering algorithms and choose the best one for this dataset. A Principal Component - weighted Euclidean distance similarity metric was computed using three existing miscible CO2 flooding projects (Weyburn, Hansford Marmaton, and Paradis) as test cases to confirm the clustering results. The clustering analysis identified five different classes of miscible CO2 projects, each with its reservoir and fluid characteristics. Type 1 projects, in general, are those that are carried out primarily in shallow carbonate reservoirs at the lowest temperatures and pressures of any database project with typical porosity and permeability. More than 60% of Type 2 projects are in sandstone reservoirs. They are at shallower depths with lower temperatures, pressures, porosities, and permeabilities than project Type 4. Type 3 projects are typically undertaken in carbonate reservoirs with medium depths and temperatures but the highest reservoir pressures. This project type has medium porosity and permeability. In comparison to project Type 2, Type 4 primarily consists of projects conducted in sandstone formations at great depths with high reservoir temperatures and pressures. The porosity and permeability of these project types are average. Finally, Type 5 projects are typically undertaken in sandstone formations at average depths, temperatures, and pressures, but with the greatest porosity and permeability of all project types. In addition to the clustering analysis, the distance similarity metric used in this work identified projects that were most like the test miscible CO2 flooding project cases. Key rock and fluid properties, well types, and best infill drilling strategies, recovery improvement strategies, and production performance can all be learned from identified analogue projects. This data can be used to improve the operational, technical, field, and well-planning decisions for new CO2 flooding projects. The workflow demonstrated in this paper is easily adaptable to data sets from other flooding projects.

Computer tomography technology is widely used in geological exploration because it is a nondestructive and three-dimensional imaging method that can be integrated with computer simulation. However, the large-scale application of the computer tomography technique is limited by economic costs and time consumption. Therefore, it is challenging and intractable to indicate the pore structure characteristics of rock. To address this issue, a super-resolution reconstruction algorithm based on convolutional neural networks, residual learning, and attention mechanism was proposed to generate super-resolution images in this study. This algorithm was applied to the reconstruction of carbonate rock and sandstone. The performance of two-dimensional image reconstruction was evaluated by quantitative extraction and qualitative visualization. The results from experiments indicate that the built model performs well on different upscaling factors and is superior to the existing super-resolution approaches based on convolutional neural network.