Actual Line Research Articles

Current coordinated distribution and voltage optimization control for multi-VSGs system considering nonlinear and negative sequence loads

In a microgrid system where multiple virtual synchronous generators (VSGs) are interconnected through high impedance cables, the total output negative sequence and harmonic current of multiple VSGs cannot be coordinated based on the capacity of each power conversion system (PCS). In addition, there is significant distortion in the voltage at the point of common coupling (PCC). Because the impedance amplitude and impedance angle of the cable change with the harmonic frequency, the resistive virtual impedance reshaping method and fixed impedance angle virtual impedance reshaping method used in the current research cannot solve the harmonic current distribution problem. To solve the above problems, a multi-frequency points VSG impedance reshaping control strategy is proposed in this paper, which can adjust the resistive and inductive output impedance of each harmonic frequency independently. The strategy is based on the harmonic separation algorithm of LPF with Feedforward Compensation (FCLPF) and the voltage controller of Vector Proportional Integral (VPI) in fundamental dq rotation coordinate system, which effectively improve the flexibility and accuracy of harmonic impedance reshaping. Among them, FCLPF harmonic separation algorithm has the characteristics of target frequency bandpass and non-target frequency notch, and can eliminate the coupling interference of each harmonic virtual impedance when accumulating virtual impedance voltage references at various frequency points. The closed-loop transfer function of the multi-parallel VPI voltage controller has the characteristics of unit gain and zero phase shift at each harmonic frequency point, which means the static response error of virtual impedance voltage command can be almost completely eliminated. Under the proposed control strategy, the fundamental positive power can be allocated according to the droop coefficient, the fundamental negative current and harmonic current can be allocated according to the sum of the reshaped virtual impedance and the actual line impedance. Reshaping virtual impedance at each harmonic frequency to negative value can also compensate the harmonic voltage drop on the high impedance cable and improve the voltage quality at the PCC.

Cyclic Air Braking Strategy for Heavy Haul Trains on Long Downhill Sections Based on Q-Learning Algorithm

Cyclic air braking is a key factor affecting the safe operation of trains on long downhill sections. However, a train’s cycle braking strategy is constrained by multiple factors such as driving environment, speed, and air-refilling time. A Q-learning algorithm-based cyclic braking strategy for a heavy haul train on long downhill sections is proposed to address this challenge. First, the operating environment of a heavy haul train on long downhill sections is designed, considering various constraint parameters, such as the characteristics of special operating routes, allowable operating speeds, and train tube air-refilling time. Second, the operating status and braking operation of a heavy haul train on long downhill sections are discretized in order to establish a Q-table based on state–action pairs. The training of algorithm performance is achieved by continuously updating Q-tables. Finally, taking the heavy haul train formation as the study object, actual line data from the Shuozhou–Huanghua Railway are used for experimental simulation, and different hyperparameters and entry speed conditions are considered. The results show that the safe and stable cyclic braking of a heavy haul train on long downhill sections is achieved. The effectiveness of the Q-learning control strategy is verified.

Real-time detection of surface cracking defects for large-sized stamped parts

This study presents a framework for the real-time detection of surface cracking in large-sized stamped metal parts. The framework aims to address the challenges of low detection efficiency and high error rates associated with manual cracking detection. Within this framework, a novel network, SNF-YOLOv8, is proposed to efficiently detect cracking while ensuring that the detection speed matches the production speed. The network incorporates a convolutional spatial-to-depth module to enhance the detection of small-sized cracking and mitigate surface interference during inspections. Furthermore, a visual self-attention mechanism is introduced to improve feature extraction. A combination of standard convolutional and depth-wise separable convolutional layers in the neck network enhances speed without compromising accuracy. Experimental validation conducted using a dataset from actual production lines, in collaboration with a multi-national corporation, demonstrates that SNF-YOLOv8 achieves an average precision of 85.2% at a detection speed of 164 frames per second. The framework achieves an accuracy rate of 98.8% in detecting large-sized cracking and 96.4% in detecting small-sized cracking, meeting the requirements for high-precision and real-time detection applications.

Preoperative phenotype has no significant impact on the clinical outcomes and long-term survival of mechanically aligned total knee arthroplasty in Asian patients with osteoarthritis.

The aim of this study was to investigate the distribution of phenotypes in Asian patients with end-stage osteoarthritis (OA) and assess whether the phenotype affected the clinical outcome and survival of mechanically aligned total knee arthroplasty (TKA). We also compared the survival of the group in which the phenotype unintentionally remained unchanged with those in which it was corrected to neutral. The study involved 945 TKAs, which were performed in 641 patients with primary OA, between January 2000 and January 2009. These were classified into 12 phenotypes based on the combined assessment of four categories of the arithmetic hip-knee-ankle angle and three categories of actual joint line obliquity. The rates of survival were analyzed using Kaplan-Meier methods and the log-rank test. The Hospital for Special Surgery score and survival of each phenotype were compared with those of the reference phenotype with neutral alignment and a parallel joint line. We also compared long-term survival between the unchanged phenotype group and the corrected to neutral alignment-parallel joint line group in patients with Type IV-b (mild to moderate varus alignment-parallel joint line) phenotype. The most common phenotype was Type I-b (mild to moderate varus alignment-medial joint line; 27.1% (n = 256)), followed by Type IV-b (23.2%; n = 219). There was no significant difference in the clinical outcomes and long-term survival between the groups. In Type IV-b phenotypes, the neutrally corrected group showed higher 15-year survival compared with the unchanged-phenotype group (94.9% (95% confidence interval (CI) 92.0 to 97.8) vs 74.2% (95% CI 98.0 to 100); p = 0.020). Constitutional varus was confirmed in more than half of these patients. Mechanically aligned TKA can achieve consistent clinical outcomes and long-term survival, regardless of the patient's phenotype. The neutrally corrected group had better long-term survival compared with the unchanged phenotype group.

A method for calculating space charge distribution in ultra-high voltage direct current transmission lines based on the transport of diluted species theory

The occurrence of corona discharge in ultra-high-voltage direct current transmission lines results in the generation of space charge, which significantly affects the nearby electromagnetic environment. To investigate the micro-physical mechanisms of discharge and provide better guidance for the calculation of the ground ion current density and total electric field, a method based on the theory of the transport of diluted species is proposed. This method enables the visualization and analysis of the movement process of space charge in air medium, as well as the calculation of the total electric field below the transmission lines. To validate the effectiveness of this algorithm, experimental tests of the total electric field are conducted on actual transmission lines. The experimental results demonstrate that the algorithm not only highlights the evident differences in the transient microscopic processes between positive and negative corona discharges but also accurately calculates the total electric field below the transmission lines. In addition, considering the weather conditions in the Chongqing region, relevant recommendations are provided under the influence of wind speed, aiming to offer protection guidelines for individuals engaged in activities near transmission lines.

Delay analysis of ring resonator-based beam-forming network in z-domain

During the next generation of wireless cellular networks, the millimeter-wave (mm-wave) spectrum will bring new opportunities for exceptionally high data transfer speeds and extensive network connectivity. Millimeter waves, on the other hand, are subject to a significant loss of propagation, which is the most significant impediment. A beneficial solution to this difficulty, which can be overcome, is to use a beam-forming system that consists of many antennas. The purpose of this study is to provide a concept for an integrated photonic beam-forming system that utilises multiple ring resonators for a 1 × 4 phase array antenna operating in the Ka-Band frequency range. The waveguide technology is the foundation for a signal that operates at 28 GHz. It is through the use of the optical ring resonator that the actual time delay line may accomplish its goal. The suggested method can be imDuring the next generation of wireless cellular networks, the millimeter-wave (mm-wave) spectrum will bring new opportunities for exceptionally high data transfer speeds and extensive network connectivity. Millimeter waves, on the other hand, are subject to a significant loss of propagation, which is the most significant impediment. A beneficial solution to this difficulty, which can be overcome, is to use a beam-forming system that consists of many antennas. The purpose of this study is to provide a concept for an integrated photonic beam-forming system that utilises multiple ring resonators for a 1 × 4 phase array antenna operating in the Ka-Band frequency range. The waveguide technology is the foundation for a signal that operates at 28 GHz. It is through the use of the optical ring resonator that the actual time delay line may accomplish its goal. The suggested method can be implemented as a variable true time delay (TTD) line to change the radiation angle of phase array antennas (PAA). The main lobe radiated by the PAA can be directed squint-free between the angles from −28° to +28°. The mathematical analysis and design of the beam producing the structure are presented. Following that, delays of 650 ps, 350 ps, and 250 ps could be produced with coupling coefficients of κ = 0.5 , κ = 0.7, and κ = 0.9 , respectively, and the associated phase shifts were 0.469π, 0.146π, and 0.387π.plemented as a variable true time delay (TTD) line to change the radiation angle of phase array antennas (PAA). The main lobe radiated by the PAA can be directed squint-free between the angles from −28° to +28°. The mathematical analysis and design of the beam producing the structure are presented. Following that, delays of 650 ps, 350 ps, and 250 ps could be produced with coupling coefficients of κ = 0.5 , κ = 0.7, and κ = 0.9 , respectively, and the associated phase shifts were 0.469π, 0.146π, and 0.387π.

Influence of flange deflection wear and the economy-oriented reversing control policy

In this paper, we describe the phenomenon of deflection wear of the wheel flange and analyze the causes according to actual line conditions, and the influence of deflection wear of wheel flange is analyzed from the perspective of safety and economy of trains by combining dynamic modeling and simulation. The degree of deflection wear of wheel flange is divided into different grades. Based on the reliability-cost function, the influence of economy is quantified and introduced into the analysis of deflection wear as an index. We establish an economic model of flange deflection wear and cutting amount of reprofiling (FDW-CA). Based on the deflection wear prediction model and FDW-CA model, we model the economy-oriented reversing control policy. To verify the feasibility and validity of the control policy. The measured data before and after the reversing policy are analyzed. The results indicate that the cutting amount of wheelset is reduced from 11.2 to 6.6 mm on average for single reprofiling, and the saving is more than 40%. The reversing control policy can reduce the cutting amount of wheelset, reduce the limit probability of wheelset, extend the life of the wheelset, and reduce the operation and maintenance cost of the wheelset, and further provide basis for wheelset spare planning.

Full-Section Deformation Monitoring of High-Altitude Fault Tunnels Based on Three-Dimensional Laser Scanning Technology.

In traditional tunnel monitoring, the characteristic points of an object within a tunnel are measured to obtain information about the object. Considering the limitations of the traditional method in measuring the complex surface structure of tunnels, such as limited monitoring points, a long measurement period, and low precision, this study introduces an approach that uses three-dimensional (3D) laser scanning for monitoring tunnel cross-section deformation. Using this approach, the soft surrounding rock of a high-altitude ultralong tunnel was taken as the monitoring object. The test tunnel was first scanned using a 3D laser scanner, and the collected data were processed. The internal structural data of the tunnel were subsequently compared with its actual contour lines and the data of its primary branch and secondary lining on different dates. The results indicate that the arch roof of the tunnel tended to be stable within a certain time range when the positions of the primary branch and secondary lining were at different measuring points with different pile numbers. The deformation of the pile number on the left and right sides did not generally exceed 0.02 m, except at a few measuring points. A comparison between the actual cross section of the initial branch and that of the designed section showed that the actual elevation of the arch of the initial branch of the tunnel was greater than its designed elevation by no more than 0.3 m. Hence, through this study, a convenient and practical method is presented for monitoring deformation in complex curved tunnel structures.

A model for simulating PM2.5 concentration in subway systems

Due to adverse effects on health, the PM2.5 of subway systems is increasingly attracting people's attention. Although scholars have conducted many experimental studies on the characteristics of PM2.5 in subway stations, theoretical research on the propagation law of PM2.5 in subway systems is still lacking. In this study, we established a mathematical model for the PM2.5 concentration in the subway system based on the principle of mass conservation, including a PM2.5 generation model for trains operating in tunnels and a network model for calculating the ventilation rate, to simulate the variation of PM2.5 concentration in various areas of the subway system. The model was used to simulate the PM2.5 concentration of a subway line in Nanjing, China and on-site testing was conducted to verify the simulation results, with an error was within 10%. The simulation results quantitatively illustrate the influence of train density and ambient concentrations on PM2.5 variation in the subway system and the comparison with studies in other countries has also verified the applicability. This model could be used based on actual subway lines to simulate the PM2.5 concentration at any time during subway operation, thereby providing effective guidance for improving the environmental air quality of subway stations and optimizing the design of environmental control systems.

Experimental Investigation of a Rapid Calculation and Damage Diagnosis of the Quasistatic Influence Line of a Self-Anchored Suspension Bridge Based on Deflection Theory

AbstractBased on deflection theory, the basic differential equations and deformation compatibility equations of self-anchored suspension bridges are established. The calculation formula for the stiffening girder deflection of a three-span self-anchored suspension bridge is derived, and the validity of the formula is confirmed through an example. In addition, a fast calculation method for the theoretical deflection influence line is proposed by using the deflection calculation formula, and a calculation method for the actual deflection influence line is founded on the concept of time and space coordinate transformation. By loading the bridge with a load car and using real-time monitoring of the displacement response of the bridge test section, combined with the proposed intelligent damage evaluation program, the abnormal situation of the damage of the three-span self-anchored suspension bridge can be quickly and accurately identified. The research results of this paper can provide flexible, economical, and effective guidance for detecting the capacity of real bridges.

Effects of camera external parameters error on measurement accuracy in monocular vision

Monocular vision is widely employed in industrial measurement scenarios due to its low cost, flexibility, mature algorithms, and high reliability. However, in actual production line monocular vision measurement systems, the measured plane may be affected by motion mechanisms or device vibrations, causing it to deviate from the system calibration plane and resulting in measurement errors. Current research primarily focuses on analyzing the influence of camera internal parameters and feature extraction algorithms on measurement errors, with limited quantitative analysis of the impact of camera external parameters. To address this issue, our paper derives a mathematical expression for measurement error when rotational and translational motion occurs between the measurement and calibration planes. Theoretical analysis reveals that when the camera optical axis is perpendicular to the measurement plane, measurement error is independent of the main point coordinates of the camera’s optical axis, translation component in the measurement plane, and rotation angle around the camera’s optical axis. Instead, it exhibits a linear relationship with the translation component in the direction of the camera’s optical axis and is inversely proportional to the normalized focal length of the camera. When the optical axis of the camera is not perpendicular to the calibration plane, the measurement error exhibits a nonlinear relationship with camera’s internal and external parameters. To verify our theoretical derivation’s accuracy, we conduct measurement experiments using both industrial and telecentric lenses and analyzed rotation and translation of the measured plane separately. The measurement data was evaluated by Root Mean Square Error and Pearson Correlation Coefficient, which demonstrates that experimental and theoretical function images exhibit consistent trends, verifying our theoretical derivation’s correctness and providing a theoretical guide for improving actual visual measurement accuracy.

Balancing the Efficiency and Sensitivity of Defect Inspection of Non-Patterned Wafers with TDI-Based Dark-Field Scattering Microscopy.

In semiconductor manufacturing, defect inspection in non-patterned wafer production lines is essential to ensure high-quality integrated circuits. However, in actual production lines, achieving both high efficiency and high sensitivity at the same time is a significant challenge due to their mutual constraints. To achieve a reasonable trade-off between detection efficiency and sensitivity, this paper integrates the time delay integration (TDI) technology into dark-field microscopy. The TDI image sensor is utilized instead of a photomultiplier tube to realize multi-point simultaneous scanning. Experiments illustrate that the increase in the number of TDI stages and reduction in the column fixed pattern noise effectively improve the signal-to-noise ratio of particle defects without sacrificing the detecting efficiency.

A Robust Texture-Less Gray-Scale Surface Matching Method Applied to a Liquid Crystal Display TV Diffuser Plate Assembly System

In most liquid crystal display (LCD) backlight modules (BLMs), diffuser plates (DPs) play the essential role in blurring the backlight. A common BLM consists of multiple superimposed optical films. In a vision-based automated assembly system, to ensure sufficient accuracy, each of the multiple cameras usually shoots a local corner of the DP and jointly estimates the target pose, guiding the robot to assemble the DP on the BLM. In general, DPs are typical of texture-less objects with simple shapes. Due to the image background of superimposed multilayer optical films, the robustness of the most common detection methods must be improved to meet industrial needs. To solve the above problem, a texture-less surface matching method based on gray-scale images was proposed. An augmented and normalized gray-scale vector represents the texture-less gray-scale surface in a low-dimensional space. The cosine distance is then used to calculate the similarity between the template and matching vectors, combined with shape-based matching (SBM); the proposed method can obtain high robustness when detecting DPs. An image database from actual production lines was used in the experiment. In comparative tests with the NCC, SBM, YOLOv5s, and YOLOv5x methods, our proposed method had the best precision at all confidence thresholds. Although recall was slightly inferior to SBM, the comprehensive evaluation F1-Score reached 0.826, significantly outperforming the other methods. Regarding localizing accuracy, our algorithm also performed best, reaching 5.7 pixels. Although the time consumption of a single prediction is about 0.6 s, it can still meet industrial needs. These experimental results show that the proposed method has high robustness in detecting DPs and is especially suitable for vision-based automatic assembly tasks in BLM.

Hybrid actor-critic algorithm for quantum reinforcement learning at CERN beam lines

Free energy-based reinforcement learning (FERL) with clamped quantum Boltzmann machines (QBM) was shown to significantly improve the learning efficiency compared to classical Q-learning with the restriction, however, to discrete state-action space environments. In this paper, the FERL approach is extended to multi-dimensional continuous state-action space environments to open the doors for a broader range of real-world applications. First, free energy-based Q-learning is studied for discrete action spaces, but continuous state spaces and the impact of experience replay on sample efficiency is assessed. In a second step, a hybrid actor-critic (A-C) scheme for continuous state-action spaces is developed based on the deep deterministic policy gradient algorithm combining a classical actor network with a QBM-based critic. The results obtained with quantum annealing (QA), both simulated and with D-Wave QA hardware, are discussed, and the performance is compared to classical reinforcement learning methods. The environments used throughout represent existing particle accelerator beam lines at the European Organisation for Nuclear Research. Among others, the hybrid A-C agent is evaluated on the actual electron beam line of the Advanced Wakefield Experiment (AWAKE).

Spiral-wound organic solvent membrane modules for dewaxing solvent recovery

The petrochemical industry is typically energy-intensive, involving frequent and intricate separation processes. SINOPEC is a globally recognized lubricant manufacturer and distributor, with a substantial production capacity exceeding 1.7 million tonnes in 2022. The dewaxing solvent recovery process in lubricant production necessitates a multi-stage flash and distillation treatment, resulting in substantial energy consumption and high costs. Utilizing organic solvent nanofiltration (OSN) membranes offers a feasible alternative approach to address this issue. A series of thin film composite OSN membranes with varying molecule weight cut-offs were prepared through precise adjustment of the chemical composition and proportion of the organic co-monomers. The application potential of the optimized membranes in the lubricant dewaxing solvent recovery was evaluated using lubricant base oil derived from the actual production line, specifically light deasphalted oil and vacuum cut2. The membrane exhibited a rejection of 99.3 % for light deasphalted oil and 95.6 % for vacuum cut2, with corresponding fluxes of 16.8 and 13.4 LMH under 2 MPa, respectively. Based on this, spiral-wound membrane modules (1.8″ × 12″) were fabricated and their suitability for dewaxing solvent recovery was further validated. The fabrication of spiral-wound membrane modules (8″ × 40″) is in progress, with intentions to conduct a pilot plant trial on the lubricant oil manufacturing line.

Improving Recognition of Defective Epoxy Images in Integrated Circuit Manufacturing by Data Augmentation

This paper discusses the problem of recognizing defective epoxy drop images for the purpose of performing vision-based die attachment inspection in integrated circuit (IC) manufacturing based on deep neural networks. Two supervised and two unsupervised recognition models are considered. The supervised models examined are an autoencoder (AE) network together with a multi-layer perceptron network (MLP) and a VGG16 network, while the unsupervised models examined are an autoencoder (AE) network together with k-means clustering and a VGG16 network together with k-means clustering. Since in practice very few defective epoxy drop images are available on an actual IC production line, the emphasis in this paper is placed on the impact of data augmentation on the recognition outcome. The data augmentation is achieved by generating synthesized defective epoxy drop images via our previously developed enhanced loss function CycleGAN generative network. The experimental results indicate that when using data augmentation, the supervised and unsupervised models of VGG16 generate perfect or near perfect accuracies for recognition of defective epoxy drop images for the dataset examined. More specifically, for the supervised models of AE+MLP and VGG16, the recognition accuracy is improved by 47% and 1%, respectively, and for the unsupervised models of AE+Kmeans and VGG+Kmeans, the recognition accuracy is improved by 37% and 15%, respectively, due to the data augmentation.

One-dimensional parallel forward modeling for geophysical electromagnetic fields excited by sagging overhead transmission lines

Geophysicists often regard the 50/60 Hz electromagnetic (EM) waves and their harmonics, excited by overhead transmission lines, as noise that interferes with geophysical prospecting. However, some researchers view them as effective signals for geophysical exploration. Forward modeling can be used to study the characteristics of EM fields generated by overhead transmission lines. It can help in either eliminating or utilizing power-frequency EM fields. Nevertheless, current research on the forward modeling of EM fields excited by transmission lines neglects the sagging shape and direction of the actual transmission lines, simplifying them into infinitely long straight lines. In this research, we develop a one-dimensional forward modeling algorithm specifically designed for overhead transmission lines, which can accurately calculate the EM fields excited by realistic overhead transmission lines considering the three-phase system, design parameters, sag, and orientation of the transmission lines. To construct an accurate transmission line model, we represent the sagging shape of a single transmission line within a span by a catenary equation. We then derive analytical expressions for the EM fields excited by sagging overhead transmission lines. During forward modeling, an appropriate number of Gauss points must be selected to balance accuracy and speed. The optimal number of Gauss points is determined based on the distance between the observation point and the transmission line source, as well as the type of fields being calculated. OpenMP parallelization is used to improve computational speed. The forward modeling results reveal significant differences between the forward responses of a single infinitely long straight line and realistic transmission lines. This highlights the importance of fully incorporating the three-phase system, design parameters, and geometric shapes of transmission lines in the forward modeling process. Moreover, the inversion test demonstrates the potential utility of EM fields radiated by transmission lines for geophysical exploration.

Evaluation of the Wire-Guard Arrangements in Overhead Distribution Feeders Against Lightning.

This paper analyses the influence of wire-guard in distribution power systems subjected to a lightning surge. A real feeder was considered as a case study using ATP-EMTP software. The study compares the performance of the actual unprotected distribution line with a proposed shielded configuration. The simulation results show that the wire-guard can improve the immunity and power quality of distribution system against lightning.

Development of Rotor Balancing Algorithm for a High-Precision Rotor System considering Dynamic Reliability through Automatic-Adaptive DBSCAN

Recently, the demand for high-precision balancing of rotors has increased in the automobile industry, as more rotors are designed to rotate at ever-higher speeds to maximize energy efficiency. The accumulation of measurement uncertainty in the balancing process decreases the accuracy of the unbalanced mass estimation, which is the ultimate goal of balancing. Here, the problem of uncertainty is shown through a Monte Carlo simulation of signals acquired from an actual production line. To reduce the effect of measurement uncertainty in the balancing procedures, a signal-processing technique that increases the dynamic reliability of the signal is proposed. The suggested method is based on density-based spatial clustering of applications with noise (DBSCAN) with the use of the orthogonality-based averaging method. Specifically, by adjusting radius values while clustering samples through the use of the DBSCAN method, the outliers that arise due to uncertainty are successfully removed. In this work, our proposed automatic-adaptive DBSCAN (AA-DBSCAN) method is validated by applying it to a balancing machine used for blower rotors in fuel cell electric vehicles. The results show that the deviation of the extracted influence coefficients is up to 0.0050, whereas the proposed method reduced it to less than 0.0037. In addition, the suggested procedure reduced the deviations of the unbalanced mass phase estimation by 35.2% as compared to the results found by the conventional method. Consequently, through the validation test, the suggested method was found to have the largest vibration decrease of any method considered in the study.

OrangeStereo: A navel orange stereo matching network for 3D surface reconstruction

Three-dimensional (3D) surface reconstruction is crucial to accurately detecting navel orange external qualities such as size and shape, and binocular stereo vision is promising for practical sorting scenarios due to its high frame rates and ease of deployment. For objects with texture-less surface, such as navel oranges, it is very challenging to match the corresponding points using traditional stereo matching algorithms because the detected feature points have high similarity in color as well as structure. Most of the existing deep learning-based stereo matching algorithms focus on the stereo matching performance of the whole image and lack the attention to the matching performance of the main region. Meanwhile, simple background with less texture will bring more uncertainty to the matching of corresponding points, which affects the robustness of the model. The OrangeStereo was proposed based on the GwcNet structure to achieve a better corresponding points matching effect of navel oranges, which can provide more accurate 3D surface information. The Structural Features Extraction (SFE) module was introduced after the initial feature extraction module to extract stable structural features and avoid the limitations of using single convolutional features. The Attention Weights Generation (AWG) module was utilized at matching cost calculation stage to help suppress redundant information and make the network pay more attention to the matching performance of specific regions. The improved loss function assigned a greater weight to the loss of the navel orange region using semantic information from left image to make the network more focused on the matching effect of the navel orange region. An ablation study was conducted to confirm the efficiency of the SFE module, AWG module, and the improved loss function of the algorithm. The ablation study contained five sub-experiments, including the baseline model, adding the SFE and the AWG modules respectively, and whether to add the improved loss function based on the addition of previous two modules. Additionally, PSMNet, GwcNet, RealtimeStereo, and ACVNet, four typical stereo matching methods, were contrasted with the OrangeStereo. The proposed OrangeStereo achieved better performance with EPE, RMSE, bad-1, bad-3, bad-5, and inference time of 0.60 pixels, 3.53 pixels, 2.37 %, 1.83 %, 1.62 %, and 33 ms in the orange region, respectively. The R2 and RMSE for the depth of orange utilizing our method were 0.982 and 0.81 mm, respectively. The proposed algorithm can be employed to obtain accurate depth information of fruits quickly, and it is expected to be used in the 3D reconstruction of fruits in the actual commercial fruit sorting lines to achieve more accurate external quality assessment.