Vision measurement of gear pitting based on DCGAN and U-Net

Zhiwen Wang,Yi Qin,Weiwei Chen 대한기계학회 2021 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.35 No.7

To quantitatively detect the gear pitting, this paper proposes a vision measurement method based on deep convolutional generative adversarial network (DCGAN) and fully convolutional segmentation network (U-Net) for measuring the area ratio of gear pitting. A machine vision system is designed to automatically collect the pitting images of all gear teeth obtained from gear fatigue tests, but the pitting images are not sufficient due to the high cost of gear fatigue test. To solve the problem of small sample, DCGAN is applied to expand pitting samples. By the expansive sample set, the edge detection algorithm and U-Net are respectively used to segment the tooth surface and pitting. The proposed approach is applied to measure the gear pitting, and a comprehensive evaluation index is designed to evaluate the performance of gear pitting detection. Experimental results show that the average relative error and the absolute error of pitting area ratio obtained by the proposed method are respectively 7.83 % and 0.18 %, which are much lower than those obtained by the conventional detection method without sample augmentation. Thus, the proposed method has satisfactory accuracy and precede the detection method without sample augmentation.

Continuous Pad Dyeing of Cotton Fabric in a Non-Aqueous Medium for Realizing Ultra-High Fixation and Washing-Free

Zhiwen Wang,Liujun Pei,Hongjuan Zhang,Hao Li,Jiping Wang 한국섬유공학회 2022 Fibers and polymers Vol.23 No.5

In traditional water-based dyeing system, a low fixation of reactive dye requires six to ten times washing afterdyeing, which consumes a lot of water and results in huge dyeing effluent. To alleviate this issue, the development of asustainable dyeing technology of cotton textile is essential. In the present investigation, silicone non-aqueous medium wasemployed to improve the fixation of dye with a pad-dyeing process under the optimized dyeing parameters. The appropriatewashing-free dyeing conditions were recommended as follows: the padded fabric was placed at 105 oC for 15-30 s tocomplete pre-drying to decrease the water content rapidly, and then fixed in non-aqueous medium for 80-90 s at 95 oC. Afterdyeing, the dyed fabric was treated with 30 g/l fixing agent through pre-dried at 80 oC and baked at 160 oC. The resultsindicated that the final fixation of dye was above 95 %, and the dry rubbing fastness of dyed cotton fabric can reach grade 5. Moreover, the wet rubbing fastness and washing fastness of dyed cotton fabric can reach grade 3-4 and 4, respectively. Therefore, it can obtain the washing-free dyeing of cotton fabric with reactive dye. If the washing procedure was omitted inthe industrial application, the environmental cost of cotton dyeing would be greatly decreased. This method provides asubstantial reduction in wastewater effluent, which will be considerably potential for industrial application.

Apoptosis induced in vivo by new type gosling viral enteritis virus

Shun Chen,Anchun Cheng,Mingshu Wang,Dekang Zhu,Renyong Jia,Qihui Luo,Hengmin Cui,Yi Zhou,Yin Wang,Zhiwen Xu,Zhengli Chen,Xiaoyue Chen,Xiaoyu Wang 대한수의학회 2011 JOURNAL OF VETERINARY SCIENCE Vol.12 No.4

In this study, apoptosis was induced by new type gosling viral enteritis virus (NGVEV) in experimentally infected goslings is reported in detail for the first time. After 3-day-old goslings were orally inoculated with a NGVEV-CN strain suspension, the time course of NGVEV effects on apoptotic morphological changes of the internal tissues was evaluated. These changes were observed by histological analysis with light microscopy and ultrastructural analysis with transmission electron microscopy. DNA fragmentation was assessed with a terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) assay and DNA ladder analysis. A series of characteristic apoptotic morphological changes including chromatin condensation and margination, cytoplasmic shrinkage, plasma membrane blebbing, and formation of apoptotic bodies were noted. Apoptosis was readily observed in the lymphoid and gastrointestinal organs, and sporadically occurred in other organs after 3 days post-infection (PI). The presence and quantity of TUNEL-positive cells increased with infection time until 9 days PI. DNA extracted from the NGVEV-infected gosling cells displayed characteristic 180~200 bp ladders. Apoptotic cells were ubiquitously distributed, especially among lymphocytes, macrophages, monocytes, and epithelial and intestinal cells. Necrosis was subsequently detected during the late NGVEV-infection phase, which was characterized by cell swelling, plasma membrane collapse, and rapidly lysis. Our results suggested that apoptosis may play an important role in the pathogenesis of NGVE disease.

Impact of axial power distribution on thermal-hydraulic characteristics for thermionic reactor

Zhiwen Dai,Chenglong Wang,Dalin Zhang,Wenxi Tian,Suizheng Qiu,G.H. Su 한국원자력학회 2021 Nuclear Engineering and Technology Vol.53 No.12

Reactor fuel's power distribution plays a vital role in designing the new generation thermionic SpaceReactor Power Systems (SRPS). In this paper, the 1/12th SPACE-R's full reactor core was numericallyanalyzed with two kinds of different axial power distribution, to identify their impacts on thermalhydraulic and thermoelectric characteristics. In the benchmark study, the maximum error betweennumerical results and existing data or design values ranged from 0.2 to 2.2%. Four main conclusions wereobtained in the numerical analysis: a) The axial power distribution has less impact on coolant temperature. b) Axial power distribution influenced the emitter temperature distribution a lot, when the corepower was cosine distributed, the maximum temperature of the emitter was 194 K higher than that ofthe uniform power distribution. c) Comparing to the cosine axial power distribution, the uniform axialpower distribution would make the maximum temperature in each component of the reactor core muchlower, reducing the requirements for core fuel material. d) Voltage and current distribution were similarto the axial electrode temperature distribution, and the axial power distribution has little effect on theoutput power

Ensemble convolutional neural networks for automatic fusion recognition of multi‐platform radar emitters

Zhiwen Zhou,Gaoming Huang,Xuebao Wang 한국전자통신연구원 2019 ETRI Journal Vol.41 No.6

Presently, the extraction of hand‐crafted features is still the dominant method in radar emitter recognition. To solve the complicated problems of selection and updation of empirical features, we present a novel automatic feature extraction structure based on deep learning. In particular, a convolutional neural network (CNN) is adopted to extract high‐level abstract representations from the time‐frequency images of emitter signals. Thus, the redundant process of designing discriminative features can be avoided. Furthermore, to address the performance degradation of a single platform, we propose the construction of an ensemble learning‐based architecture for multi‐platform fusion recognition. Experimental results indicate that the proposed algorithms are feasible and effective, and they outperform other typical feature extraction and fusion recognition methods in terms of accuracy. Moreover, the proposed structure could be extended to other prevalent ensemble learning alternatives.

Design and analysis of a free-piston stirling engine for space nuclear power reactor

Dai, Zhiwen,Wang, Chenglong,Zhang, Dalin,Tian, Wenxi,Qiu, Suizheng,Su, G.H. Korean Nuclear Society 2021 Nuclear Engineering and Technology Vol.53 No.2

The free-piston Stirling engine (FPSE) has been widely used in aerospace owing to its advantages of high efficiency, high reliability, and self-starting ability. In this paper, a 20-kW FPSE is proposed by analyzing the requirements of space nuclear power reactor. A code was developed based on an improved simple analysis method to evaluate the performance of the proposed FPSE. The code is benchmarked with experimental data, and the maximum relative error of the output power is 17.1%. Numerical results show that the output power is 21 kW, which satisfies the design requirements. The results show that: a) reducing the pressure shell's thickness can improve the output power significantly; b) the system efficiency increases with the wire porosity, while the growth of system efficiency decreases when the porosity is higher than 80%, and system efficiency exhibits a linear relationship with the temperatures of the cold and hot sides; c) the system efficiency increases with the compression ratio; the compression ratio increases by 16.7% while the system efficiency increases by 42%. This study can provide valuable theoretical support for the design and analysis of FPSEs for space nuclear power reactors.

Research on constitutive model of aluminum alloy 7075 thermal deformation based on deep neural network

Guan Wang,Pei Zhang,Linyuan Kou,Yan Wu,Tianxiang Wen,Xin Shang,Zhiwen Liu 대한기계학회 2023 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.37 No.2

The hot deformation behavior of the Al-Zn-Mg-Cu alloy was studied by isothermal tensile tests in the temperature range of 200-350 °C and the strain rate range of 0.001-0.1 s -1 . A data-driven deep neural network (DNN) constitutive model and a phenomenological Arrhenius constitutive model were developed for the studied alloy model. The parameters of the DNN model were optimized to improve the prediction accuracy of flow stress. The results show that the accuracy of predictions of the DNN model is better than the Arrhenius model for the hot deformation behavior of 7075 aluminum alloy. The average absolute relative error and correlation coefficient of the DNN model is 1.70 % and 0.9996, respectively. The accuracy of the constitutive model of Arrhenius is relatively low for 7075 aluminum alloy in the range 200-350 °C, 0.001-0.1 s -1 . The optimal network depth and the number of neurons per layer for the analytically optimized DNN constitutive model are 6 and 28, respectively. In addition, the developed DNN model can be effectively applied in intelligent manufacturing, such as short-process highefficiency hot stamping and other plastic-forming technologies.

Design and heat transfer optimization of a 1 kW free-piston stirling engine for space reactor power system

Dai, Zhiwen,Wang, Chenglong,Zhang, Dalin,Tian, Wenxi,Qiu, Suizheng,Su, G.H. Korean Nuclear Society 2021 Nuclear Engineering and Technology Vol.53 No.7

The Free-Piston Stirling engine (FPSE) is of interest for many research in aerospace due to its advantages of long operating life, higher efficiency, and zero maintenance. In this study, a 1-kW FPSE was proposed by analyzing the requirements of Space Reactor Power Systems (SRPS), of which performance was evaluated by developing a code through the Simple Analysis Method. The results of SAM showed that the critical parameters of FPSE could satisfy the designed requirements. The heater of the FPSE was designed with the copper rectangular fins to enhance heat transfer, and the parametric study of the heater was performed with Computational Fluid Dynamics (CFD) software STAR-CCM⁺. The Performance Evaluation Criteria (PEC) was used to evaluate the heat transfer enhancement of the fins in the heater. The numerical results of the CFD program showed that pressure drop and Nusselt number ratio had a linear growth with the height of fins, and PEC number decreased as the height of fins increased, and the optimum height of the fin was set as 4 mm according to the minimum heat exchange surface area. This paper can provide theoretical supports for the design and numerical analysis of an FPSE for SRPSs.

Computer simulation of the nonlinear static behavior of axially functionally graded microtube with porosity

Li, Xiaohuan,Wang, Tian,Liu, Fang,Zhu, Zhiwen Techno-Press 2021 Advances in nano research Vol.11 No.4

Static analysis of microstructures, including bending and buckling, is widely practiced in the fabrication and creation of applications such as actuation, sensing, and energy recovery. This article aims to inquire about the static behavior of non-uniform and imperfect microtubes through a numerical solution. Based on the modified couple stress theory, the first-order shear deformation theory and Von-Karman nonlinear theory, and employing the energy conservation method, the linear and nonlinear governing equations are derived. The porosity-dependent material in both ceramic and metal phases makes the functionally graded materials which are varied along tube length, moreover, cross-sections are also considered uniform and nonuniform via three valuable functions. Finally, the linear and nonlinear equations are solved utilizing the generalized differential quadrature method (GDQM) coupled with the numerical iteration method.

Three-Dimensional Porous Carbon Nanotube Papers as Current Collector and Buffer for SnO2 Anodes

Xiaogang Sun,Zhiwen Qiu,Long Chen,Manyuan Cai,Jie Wang,Xu Li,Wei Chen,Zhenhong Liu,Hao Tang 성균관대학교(자연과학캠퍼스) 성균나노과학기술원 2017 NANO Vol.12 No.11

A novel three-dimensional porous conductive papers have been successfully synthesized via a simple physical route. Multi-walled carbon nanotubes (MWCNTs)@SnO2 composite anode materials are embedded in porous conductive papers. The peculiar structure can accommodate the huge volume expansion of MWCNTs@SnO2 composite anode materials during charge–discharge process. The framework formed by MWCNTs and cellulose can greatly improve the strength, stability and flexibility of the electrode. In addition, the structure successfully prevent the aggregation of SnO2 nanoparticles and collapse of MWCNTs@SnO2 composite electrode, leading to the improvement in electrochemical utilization and stable cyclability. The samples were characterized by X-ray diffraction (XRD), field-emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM), respectively. The electrochemical properties and application were evaluated by galvanostatic discharge–charge testing and cycling voltammetry. As a result, the MWCNTs@SnO2 composite electrode showed excellent rate performance. The discharge capacity remains about 680mAh g -1 after 100 cycles at 200mA g -1, and even around 300mAh g -1 at 1000mA g -1.