Printed Circuit Board (PCB) Surface Micro Defect Detection Model Based on Residual Network with Novel Attention Mechanism

Printed Circuit Board(PCB)surface tiny defect detection is a difficult task in the integrated circuit industry,especially since the detection of tiny defects on PCB boards with large-size complex circuits has become one of the bottlenecks.To improve the performance of PCB surface tiny defects detection,a PCB tiny defects detection model based on an improved attention residual network(YOLOX-AttResNet)is proposed.First,the unsupervised clustering performance of the K-means algorithm is exploited to optimize the channel weights for subsequent operations by feeding the feature mapping into the SENet(Squeeze and Excitation Network)attention network;then the improved K-means-SENet network is fused with the directly mapped edges of the traditional ResNet network to form an augmented residual network(AttResNet);and finally,the AttResNet module is substituted for the traditional ResNet structure in the backbone feature extraction network of mainstream excellent detection models,thus improving the ability to extract small features from the backbone of the target detection network.The results of ablation experiments on a PCB surface defect dataset show that AttResNet is a reliable and efficient module.In Torify the performance of AttResNet for detecting small defects in large-size complex circuit images,a series of comparison experiments are further performed.The results show that the AttResNet module combines well with the five best existing target detection frameworks(YOLOv3,YOLOX,Faster R-CNN,TDD-Net,Cascade R-CNN),and all the combined new models have improved detection accuracy compared to the original model,which suggests that the AttResNet module proposed in this paper can help the detection model to extract target features.Among them,the YOLOX-AttResNet model proposed in this paper performs the best,with the highest accuracy of 98.45% and the detection speed of 36 FPS(Frames Per Second),which meets the accuracy and real-time requirements for the detection of tiny defects on PCB surfaces.This study can provide some new ideas for other real-time online detection tasks of tiny targets with high-resolution images.

Design and platform testing of the compact torus central fueling device for the EAST tokamak

Compact torus(CT)injection is a highly promising technique for the central fueling of future reactor-grade fusion devices since it features extremely high injection velocity and relatively high plasma mass.Recently,a CT injector for the EAST tokamak,EAST-CTI,was developed and platform-tested.In the first round of experiments conducted with low parameter settings,the maximum velocity and mass of the CT plasma were 150 km·s^(-1)and 90μg,respectively.However,the parameters obtained by EAST-CTI were still very low and were far from the requirements of a device such as EAST that has a strong magnetic field.In future,we plan to solve the spark problem that EAST-CTI currently encounters(that mainly hinders the further development of experiments)through engineering methods,and use greater power to obtain a more stable and suitable CT plasma for EAST.

Polymorphism of Transcription Factor 7-Like 2 in Obese Patients with Type 2 Diabetes Mellitus

Objective: The article is to investigate the relationship between the polymorphism of transcription factor 7-like 2 (TCF7L2) gene rs7903146 and type 2 diabetes mellitus with obesity in Chinese Han population. Methods: 216 patients with type 2 diabetes mellitus (T2DM), 92 males, 124 females, 194 normal controls, 89 males and 105 females. The polymorphism of TCF7L2 gene rs7903146 was measured by PCR-RFLP. PCR amplifier was used for amplification reaction, followed by polymorphism analysis of TCF7L2 gene: the reaction system was 20 μl, among which restriction endonuclease was 0.5 μl and PCR product was 7.5 μl. Restriction enzyme SspI (Fermentas) was used for PCR product and was bathed at 37°C for 16 h. The enzyme digestion products were detected and genotypes were identified by 2% agarose gel electrophoresis. The genotyping was repeated in 10% of the samples, and the genotyping rate was 100%. Height and weight were measured and BMI calculated. According to BMI, the experimental group was further divided into the obese group (n = 137, BMI > 25 kg/m2) and the normal body mass group (n = 79, BMI < 25 kgm2). Normal control group: BMI < 25 kgm2: FBG < 6.1 mmol/L and 2 h BG < 7.8 mmol/L after meal. Finally, SPSS18.0 software was used for analysis. Results: FBG, 2 h BG, FIN and Lg homa-ir showed significant differences between the control group and the T2DM groups (P < 0.01), while no statistical difference was found between the other parameters (P > 0.05). There was a significant difference in genotype frequency between the control group and the experimental group (χ2 = 17.382, P < 0.001), but no significant difference in allele frequency between the control group and the T2DM group (P > 0.05). Genotype frequency at BMI level was significantly different between the control group and the T2DM group (χ2 = 20.427, P = 0.001), but there was no significant difference in allele frequency distribution (P > 0.05). The TT type and TC + CC type were significantly different between the T2DM group and the control group (χ2 = 78.154, P = 0.006) and the T2DM obesity group and the control group (χ2 = 7.247, P = 0.009), but there was no significant difference between the T2DM normal body mass group and the control group (P > 0.05). In addition, there was no significant difference between TC type + TT type and CC type between the T2DM group and the control group and the T2DM obesity group and the control group (P > 0.05). Conclusion: There were significant differences in the genotype frequency and recessive inheritance pattern (TT/CC + TC) of rs7903146, suggesting that this SNP may be associated with the incidence of T2DM in obesity and insulin resistance.

Development of a compact torus injection system for the Keda Torus eXperiment

A compact torus injection system,KTX-CTI,has been developed for the planned injection experiments on the Keda Torus e Xperiment(KTX)reversed field pinch(RFP)device to investigate the physics and engineering issues associated with interaction between a compact torus(CT)and RFP.The key interests include fueling directly into the reactor center,confinement improvement,and the injection of momentum and helicity into the RFP discharges.The CT velocity and mass have been measured using a multichannel optical fiber interferometer,and for the first time the time evolution of the CT density profile during CT propagation is obtained.The effects of discharge parameters on the number of injected particles,CT velocity and CT density have been characterized:the maximum hydrogen CT plasma mass,m,CTis 50μg,corresponding to 30%of the mass in a typical KTX plasma;the CT velocity exceeds 120 km s-1.It is observed for the first time that multiple CTs can be produced and emitted during a very short period(<100μs)in one discharge,which is significant for the future study of repetitive CT injection,even with an ultra-high frequency.

High-energy-density plasma in femtosecond-laser-irradiated nanowire-array targets for nuclear reactions

In this work,the high-energy-density plasmas(HEDP)evolved from joule-class-femtosecond-laser-irradiated nanowire-array(NWA)targets were numerically and experimentally studied.The results of particle-in-cell simulations indicate that ions accelerated in the sheath field around the surfaces of the nanowires are eventually confined in a plasma,contributing most to the high energy densities.The protons emitted from the front surfaces of the NWA targets provide rich information about the interactions that occur.We give the electron and ion energy densities for broad target parameter ranges.The ion energy densities from NWA targets were found to be an order of magnitude higher than those from planar targets,and the volume of the HEDP was several-fold greater.At optimal target parameters,8%of the laser energy can be converted to confined protons,and this results in ion energy densities at the GJ/cm^(3) level.In the experiments,the measured energy of the emitted protons reached 4 MeV,and the changes in energy with the NWA’s parameters were found to fit the simulation results well.Experimental measurements of neutrons from 2H(d,n)3He fusion with a yield of(24±18)×10^(6)/J from deuterated polyethylene NWA targets also confirmed these results.

Investigation of the compact torus plasma motion in the KTX-CTI device based on circuit analyses

Compact torus(CT)injection is one of the most promising methods for the central fuelling of next-generation reactor-grade fusion devices due to its high density,high velocity,and selfcontained magnetised structure.A newly compact torus injector(CTI)device in Keda Torus e Xperiment(KTX),named KTX-CTI,was successfully developed and tested at the University of Science and Technology in China.In this study,first,we briefly introduce the basic principles and structure of KTX-CTI,and then,present an accurate circuit model that relies on nonlinear regression analysis(NRA)for studying the current waveform of the formation region.The current waveform,displacement,and velocity of CT plasma in the acceleration region are calculated using this NRA-based one-dimensional point model.The model results were in good agreement with the experiments.The next-step upgrading reference scheme of the KTX-CTI device is preliminarily investigated using this NRA-based point model.This research can provide insights for the development of experiments and future upgrades of the device.

Estimation of China Fusion Engineering Test Reactor performance and burning fraction in different pellet fueling scenarios by a multi-species radial transport model

Tritium self-sufficiency in future deuterium–tritium fusion reactors is a crucial challenge.As an engineering test reactor,the China Fusion Engineering Test Reactor requires a burning fraction of 3%for the goal to test the accessibility to the future fusion plant.To self-consistently simulate burning plasmas with profile changes in pellet injection scenarios and to estimate the corresponding burning fraction,a one-dimensional multi-species radial transport model is developed in the BOUT++framework.Several pellet-fueling scenarios are then tested in the model.The results show that the increased fueling depth improves the burning fraction by particle confinement improvement and fusion power increase.Nevertheless,by increasing the depth,the pellet cooling-down may significantly lower the temperature in the core region.Taking the density perturbation into consideration,the reasonable parameters of the fueling scenario in these simulations are estimated as pellet radius r_(p)=3 mm,injection rate=4 Hz,and pellet injection velocity=1000–2000 m s^(-1) without drift or 450 m s^(-1) with high-field-side drift.

Synchronous post-acceleration of laser-driven protons in helical coil targets by controlling the current dispersion

Post-acceleration of protons in helical coil targets driven by intense,ultrashort laser pulses can enhance ion energy by utilizing the transient current from the targets’self-discharge.The acceleration length of protons can exceed a few millimeters,and the acceleration gradient is of the order of GeV/m.How to ensure the synchronization between the accelerating electric field and the protons is a crucial problem for efficient post-acceleration.In this paper,we study how the electric field mismatch induced by current dispersion affects the synchronous acceleration of protons.We propose a scheme using a two-stage helical coil to control the current dispersion.With optimized parameters,the energy gain of protons is increased by four times.Proton energy is expected to reach 45 MeV using a hundreds-of-terawatts laser,or more than 100 MeV using a petawatt laser,by controlling the current dispersion.

Fabrication of large-area uniform carbon nanotube foams as near-critical-density targets for laser–plasma experiments

Carbon nanotube foams(CNFs)have been successfully used as near-critical-density targets in the laser-driven acceleration of high-energy ions and electrons.Here we report the recent advances in the fabrication technique of such targets.With the further developed floating catalyst chemical vapor deposition(FCCVD)method,large-area(>25 cm^(2))and highly uniform CNFs are successfully deposited on nanometer-thin metal or plastic foils as double-layer targets.The density and thickness of the CNF can be controlled in the range of 1−13 mg/cm^(3)and 10−200µm,respectively,by varying the synthesis parameters.The dependence of the target properties on the synthesis parameters and the details of the target characterization methods are presented for the first time.