Loading...

Table of Content

    30 June 2021, Volume 27 Issue 3
    Invited Review
    Key scientific and technological principles of hydrogen energy and fuel cells: challenges and prospects
    ZHU Mingyuan, LIU Wenbo, LIU Yang, QI Cai, LI Ying, LI Wenxian, ZHANG Jiujun
    2021, 27(3):  411-443.  doi:10.12066/j.issn.1007-2861.2300
    Asbtract ( 3393 )   HTML ( 350)   PDF (9178KB) ( 842 )  
    Figures and Tables | References | Related Articles | Metrics

    Hydrogen is a clean and sustainable secondary energy source. Its industrial chain consists of production, storage, transportation, and usage. Fuel cells, the devices that are the most efficient in terms of hydrogen use, play a pivotal role in the hydrogen industrial chain. In this paper, several fundamental scientific technologies are discussed alongside the industrialisation of hydrogen and fuel cell production. The processes and technologies introduced in this paper include hydrogen production, hydrogen storage, hydrogen refuelling stations, fuel cell stacks, key materials, and the key component/system requirements for fuel cells stacks. In addition, potential challenges as well as recent developments of the hydrogen industrial chain in China are summarised and discussed. Furthermore, to facilitate further research and development of these critical processes, several directions for future research are proposed.

    Research Articles
    Multiphysics model and numerical simulations of lead-acid battery
    SHI Meihua, DONG Li, YUAN Jingchao, ZHANG Shuxiang, SHAO Qinsi, YAN Wei, LI Jiang, LI Aijun, ZHANG Jiujun
    2021, 27(3):  444-453.  doi:10.12066/j.issn.1007-2861.2144
    Asbtract ( 1491 )   HTML ( 31)   PDF (1983KB) ( 298 )  
    Figures and Tables | References | Related Articles | Metrics

    Herein, the discharge mechanism of lead-acid batteries was discussed and a multiphysics model was proposed to simulate the battery discharge process. The model was tested and verified on the basis of the experimental data of several commercial batteries. The universality of the model was guaranteed by the average empirical parameters of the model. The model was used to analyse the structural parameters of existing batteries, and improvement of the battery structure was proposed on the basis of the simulation results. Herein, a new design scheme was presented for the low-cost fabrication of batteries having reduced volume, lighter weight, and higher capacity. Multiphysics modelling and numerical simulations could greatly reduce the cost and time required for the development of novel batteries.

    Improved approach to simultaneous left- and right-hand segmentation from a single depth image
    XU Zhengze, ZHANG Wenjun
    2021, 27(3):  454-465.  doi:10.12066/j.issn.1007-2861.2247
    Asbtract ( 1415 )   HTML ( 30)   PDF (5992KB) ( 220 )  
    Figures and Tables | References | Related Articles | Metrics

    Hand gesture recognition technology based on depth image, which relies on the accurate identification of "clean" hand in the captured depth image, is the primary interactive mode for digital media devices of future generation. We propose an improved approach to simultaneous left- and right-hand segmentation, extending the traditional SegNet algorithm by strategies including class weight, transposed convolution, hybrid dilated convolution, and skip-connection between the encoder and decoder performed by concatenation. Our approach achieves higher F2-Score than the existing baseline by 7.6% for the left and 5.9% for the right hand. The processing on the GPU reaches 20.5 ms per frame at inference time, making real-time hand tracking in depth image sequences feasible. The results of the experiment demonstrate that our approach can considerably improve the performance of simultaneous left- and right-hand segmentation from a single depth map.

    A local stereo matching algorithm based on shaped adaptive window
    LIU Jun, MIAO Zhiyong, ZHANG Yuqi, REN Jianhua
    2021, 27(3):  466-480.  doi:10.12066/j.issn.1007-2861.2255
    Asbtract ( 1580 )   HTML ( 34)   PDF (7728KB) ( 144 )  
    Figures and Tables | References | Related Articles | Metrics

    Improving the accuracy of image field depth using a stereo matching algorithm is a key challenge in the field of machine vision. A so-called shaped adaptive window local stereo matching algorithm is proposed to solve the problem that traditional adaptive window algorithms are susceptible to uneven illumination, which affects their accuracy, and that the shape of the window does not accurately describe the boundary of the image that is to be matched. Because the traditional Census transform is vulnerable to fluctuations in illumination of the central pixel, a so-called three-dimensional pixel information technique which combines with the Census transform is proposed. The matching cost is obtained by comparing the central pixel with non-central pixels and by comparing the non-central pixels with one another. Moreover, for better conformance to the image contour, and to better approximate the image boundary of the examined object to obtain higher matching accuracy by the subsequent cost aggregation operation, a double helix path method is proposed. This method obtains a window that is shaped differently from the traditional rectangle or cross-based windows. In this method, the double paths which search and examine the pixel region simultaneously shorten the time required for window construction; meanwhile, the double helix path creates a more variable window shape, which more easily conforms to the image contour. The experimental results show that, compared with most conventional stereo matching algorithms on the Middlebury platform, the proposed algorithm is better able to characterise the image boundary to improve matching precision and is more robust in cases of uneven illumination.

    Classification and recognition of underwater small targets based on improved YOLOv3 algorithm
    SHAO Huixiang, ZENG Dan
    2021, 27(3):  481-491.  doi:10.12066/j.issn.1007-2861.2279
    Asbtract ( 2231 )   HTML ( 49)   PDF (1186KB) ( 336 )  
    Figures and Tables | References | Related Articles | Metrics

    This study proposes an improved YOLOv3 algorithm designed to address the twin issues of low detection and recognition rate and high false alarm rate with respect to the detection of small targets by sonar. The improved YOLOv3 network is optimised on the basis of the original YOLOv3 algorithm, with the hierarchical connection of the network changed and the features of the shallow and deep layers fused to form a new larger-scale detection layer. Concurrently, the linear scaling $K$-means clustering algorithm is used to optimise the calculation of the number of a priori boxes and the aspect ratio, thereby improving the correlation between the a priori and ground truth boxes. These modifications improve the average accuracy of the YOLOv3 algorithm by 7%. Experimental results show that the proposed improvements to the YOLOv3 algorithm result in the effective identification of small targets with higher accuracy and lower false alarm rate, while maintaining the real-time processing capabilities of the YOLOv3 algorithm.

    Explosive synchronisation of multi-layer networks based on frequency-weighted coupled model
    JIN Yanliang, YAO Lin, WANG Xue, LUO Xuetao
    2021, 27(3):  492-502.  doi:10.12066/j.issn.1007-2861.2131
    Asbtract ( 1080 )   HTML ( 28)   PDF (875KB) ( 346 )  
    Figures and Tables | References | Related Articles | Metrics

    Synchronisation is a prevalent and universal dynamic behaviour of complex systems. Based on the critical coupling strength of explosive synchronisation in multi-layer complex networks, a frequency-weighted Kuramoto model was developed in this study. We obtained the critical coupling strength for explosive synchronisation through field tests, self-consistent analysis, and numerical simulation. The theoretical analysis and simulation results show that the critical value is related to the strength of the interlayer interaction and the average degree of the network. Increasing the interaction between layers prevents explosive synchronisation, whereas increasing the average degree of the network facilitates explosive synchronisation. The results are useful for studies on complex networks and systems.

    Modelling analysis and optimisation of high-speed signal reflection in transmission lines
    CHEN Zhangjin, WANG Wenlei, JI Yuan, HUANG Shuping
    2021, 27(3):  503-513.  doi:10.12066/j.issn.1007-2861.2171
    Asbtract ( 1495 )   HTML ( 1410)   PDF (1939KB) ( 299 )  
    Figures and Tables | References | Related Articles | Metrics

    The point-to-point transmission line is currently the most widely used and basic type of transmission line. Traditional modelling analysis methods are not fully applicable to the analysis of high-speed transmission lines. In this study, the signal reflection phenomenon in point-to-point transmission lines was modelled and analysed, and the effects of the line length and via on the signal reflection in transmission lines were analyzed. Detailed modelling of the impedance matching scheme for the signal reflection and analysis of matching resistors, branch line lengths, vias, and termination voltages in impedance matching was performed. Furthermore, the feasibility of the analytical method was verified through theoretical analysis. The analytical design scheme has been applied to the design of multiple products.

    Electroencephalogram features based on detrended fluctuation analysis of different features of music stimulation
    ZHU Jiacheng, LI Yingjie
    2021, 27(3):  514-524.  doi:10.12066/j.issn.1007-2861.2163
    Asbtract ( 1357 )   HTML ( 24)   PDF (2115KB) ( 207 )  
    Figures and Tables | References | Related Articles | Metrics

    In this study, we investigated the effects of different musical features on the electrophysiological and psychological responses of participants, and explored the long-range correlations of electroencephalograms (EEGs). We recruited 10 students to participate in four listening tasks involving different musical features. After each task, the participants completed a self-evaluation of their emotions. Scalp EEG signals were collected synchronously during the tasks. Considering the non-stationary and nonlinear characteristics of music-stimulated EEGs, we used a nonlinear method to detect the long-range correlation of non-stationary time series, namely, detrended fluctuation analysis. Long-range correlations were analysed by calculating the scale index of the EEG signal sub-band sequence, and combining it with behavioural data. The results show that the positive emotions induced by the rising tone of the happy version are significantly reduced, and whether it was rising tone or falling tone, it would significantly reduce the sad emotions induced by sad music. Under musical stimulation involving different tonal features, the subjects showed obvious brain lateralisation characteristics in the alpha and beta bands, with the left hemisphere brain dynamics showing greater activity. Moreover, the scale index used in this study was shown to reflect the specificity of EEG under different musical stimuli.

    Label-specific feature-based multi-label manifold learning
    KANG Liuyue, HUANG Rui, SUN Guangling
    2021, 27(3):  525-534.  doi:10.12066/j.issn.1007-2861.2132
    Asbtract ( 1295 )   HTML ( 30)   PDF (727KB) ( 186 )  
    Figures and Tables | References | Related Articles | Metrics

    Multi-label manifold learning (ML$^2$) constructs label manifolds based on feature manifolds and converts logical into numeric labels. This can better reflect the correlations between labels and improve classification performance. However, similar to most methods, ML$^2$ is based on all features and ignores different discriminabilities when different features are used to classify different labels. Therefore, a method we call label-specific feature-based multi-label manifold learning (LSF-ML$^2$) is proposed. First, the labels are used to optimise the feature importance matrix, which can determine the subset of label-specific features. Then, the feature manifold of the subset is mapped to the label space so that the logical labels can be converted into numeric labels. Finally, a multi-output regression is applied for classification. Experimental results show that the proposed method outperforms several existing multi-label classification methods.

    Event extraction of Chinese text based on composite neural network
    JI Zhongxiang, WU Yue
    2021, 27(3):  535-543.  doi:10.12066/j.issn.1007-2861.2223
    Asbtract ( 1318 )   HTML ( 31)   PDF (1524KB) ( 99 )  
    Figures and Tables | References | Related Articles | Metrics

    The recurrent neural network is widely used in the event extraction of Chinese text to extract events and event elements, but it usually loses essential information when processing long words. In this study, the convolutional neural network (CNN) and the bidirectional long short-term memory (Bi-LSTM) network were combined to develop a novel event extraction model known as CNN-Bi-LSTM-conditional random field (CRF). A joint vector of characters and words was adopted based on the attention mechanism and semantic features, and the CNN and Bi-LSTM models were used to process the vector to obtain its implicit representation. Finally, the CRF was used to obtain the prediction results. The experimental results show that the proposed method is more accurate than other existing event extraction methods in extracting Chinese text.

    Text classification model based on essential $n$-grams and gated recurrent neural network
    ZHAO Qian, WU Yue, LIU Zongtian
    2021, 27(3):  544-552.  doi:10.12066/j.issn.1007-2861.2158
    Asbtract ( 1233 )   HTML ( 23)   PDF (1153KB) ( 128 )  
    Figures and Tables | References | Related Articles | Metrics

    An effective text classification model based on $n$-grams and a gated recurrent neural network is proposed in this paper. First, we adopt a simpler and more efficient pooling layer to replace the traditional convolutional layer to extract the essential $n$-grams as important semantic features. Second, a bidirectional gated recurrent unit (GRU) is constructed to obtain the global dependency features of the input text. Finally, we apply the fusion model of the two features to the text classification task. We evaluate the quality of our model on sentiment and topic categorization tasks over multiple public datasets. Experimental results show that the proposed method can improve text classification effectiveness compared with the traditional model. On accuracy, it approaches an improvement of 1.95% on the 20newsgroup and 1.55% on the Rotten Tomatoes corpus.

    Improved single neuron gradient learning-based active queue management for wireless networks
    QI Aichun, XU Lei
    2021, 27(3):  553-562.  doi:10.12066/j.issn.1007-2861.2172
    Asbtract ( 1066 )   HTML ( 21)   PDF (1424KB) ( 144 )  
    Figures and Tables | References | Related Articles | Metrics

    This study introduces a scheme that takes the packet arrival link rate as the input of the controller, where the traditional network congestion control ignores the continuous state of network congestion. Then, an active queue management algorithm that improves single neuron gradient learning (ISNGL) is obtained. The algorithm uses gradient learning to adjust dynamically the network parameters and to improve the convergence rate and stability. The study also proposes a new activation function with displacement parameters and an improved method using momentum adjustment for weights. Finally, NS2 network simulation software is used to simulate a wireless network topology model. Results show that the ISNGL algorithm provides good congestion control capability in wireless networks.

    MEK5$\alpha $ and MEK5$\beta $ differentially regulate Beclin 1 promoter
    LIU Xiaoyun, WANG Leibin, WANG Qing, ZHANG Shasha, ZHAO Weimiao, HE Lin, ZHU Hongxin
    2021, 27(3):  563-572.  doi:10.12066/j.issn.1007-2861.2142
    Asbtract ( 1185 )   HTML ( 23)   PDF (2156KB) ( 79 )  
    Figures and Tables | References | Related Articles | Metrics

    Beclin 1, a mammalian autophagy-related gene, regulates autophagy initiation and autophagosome maturation. During muscle differentiation, Beclin 1 is upregulated. Additionally, MEK5-ERK5 is activated to regulate muscle differentiation. Thus, MEK5-ERK5 may regulate Beclin 1 gene expression during muscle differentiation. The aim of this study was to determine MEK5 regulation of Beclin 1 promoter in myoblast cells. A series of promoter-luciferase constructs harbouring different lengths of Beclin 1 promoter were created and transfected into myoblast C2C12 cells. In the luciferase assay, the construct containing 586 base pairs upstream of the start codon (p-354) exhibited the most potent luciferase activity. MEK5$\alpha $ and MEK5$\beta $ enhanced and suppressed p-354 luciferase activity, respectively. MEK5$\beta $ is capable of antagonizing the effect of MEK5$\alpha $ on p-354 luciferase activity. Consistent with the results on the regulatory effects of MEK5 on Beclin 1 promoter, MEK5$\alpha $CA and MEK5$\beta $DD upregulated and downregulated Beclin 1 mRNA expression, respectively. Moreover, MEK5$\beta $DD antagonised the stimulatory effects of MEK5$\alpha $CA on Beclin 1 mRNA expression. Members of the CREB family, including CREB3, CREBP, and CREBL1, promoted p-354 luciferase activity. Furthermore, CREB3 dose-dependently increased p-354 luciferase activity and exhibited a synergistic effect with MEK5$\alpha $ on p-354 luciferase activity. Collectively, these findings indicate that MEK5$\alpha $ and MEK5$\beta $ differentially regulate Beclin 1 promoter activity and that CREB family members may be downstream effectors.

    Alternative demand forecasting considering product feature attribute
    GAO Junjun, NI Ziyue
    2021, 27(3):  573-582.  doi:10.12066/j.issn.1007-2861.2173
    Asbtract ( 1174 )   HTML ( 23)   PDF (1171KB) ( 160 )  
    Figures and Tables | References | Related Articles | Metrics

    The current e-commerce sales business is developing rapidly, and the rapid and accurate prediction of demands has become a necessary research direction. The substitution of products has a significant influence on demand, and applied research in this aspect is increasing. Based on the ranking of best-selling predictive attribute values, proximity replacement rate estimation and the Adaboost prediction model were applied in this study to develop an improved demand forecasting method with higher accuracy, considering product feature attributes. The experimental findings confirm that the proposed method is accurate and reliable.

    Effect of hole opening on mechanical properties of polycrystalline plate based on crystal plastic finite element
    HU Xiaoyu, DU Yapeng, CHU Haijian
    2021, 27(3):  583-593.  doi:10.12066/j.issn.1007-2861.2159
    Asbtract ( 1599 )   HTML ( 28)   PDF (4671KB) ( 420 )  
    Figures and Tables | References | Related Articles | Metrics

    Based on the secondary development platform of ABAQUS subroutine VUMAT, the dislocation evolution and twinning mechanism have been incorporated into the crystal plastic finite element method (CPFEM), which is used to investigate the mechanical behavior of polycrystalline plastic materials. The validity of the secondary development program is verified by comparing the experimental and simulation results. The CPFEM with its twinning effect simulates and analyzes the effect of holes on the mechanical properties of the plate. Furthermore, linear approximation can be used safely when the aperture is less than half of the plate width; however, it is not suitable for larger apertures. When the spacing between two hole openings is small, their positional arrangement has a significant impact on the toughness and ultimate stress of the plate; this impact can be divided into three modes: weak influence zone, strong influence zone, and transition zone. For the plate with double holes subjected to a tensile load, an arrangement along the axis is the most optimum.

    Temperature-controlling surface plasmon resonance system and its application
    WANG Shaopeng, ZHANG Hongyan, HUANG Huaixiang
    2021, 27(3):  594-600.  doi:10.12066/j.issn.1007-2861.2180
    Asbtract ( 1566 )   HTML ( 19)   PDF (1066KB) ( 210 )  
    Figures and Tables | References | Related Articles | Metrics

    Surface plasmon resonance (SPR) sensing technology is widely used in biomedicine, pharmaceutical screening, clinical diagnosis, food safety, environmental pollution detection, and other fields. Because of the sensitivity of SPR systems to temperature, practical applications of SPR are limited. In this study, based on angle-modulation SPR, a temperature-controlling department is designed with a sensitivity of 497.8${^\circ}$/RIU (refractive index unit) and a 0.1 ℃ accuracy temperature in the range of 18$\sim$42 ℃. This SPR technology can be successfully used in various temperature changing liquid-crystal molecule systems. Results showed that the SPR signal had a good linear relationship with the temperature in the range of 25$\sim$41 ℃, with a correlation coefficient greater than 0.98.

    Estimation of mixed quantile regression parameters based on an asymmetric Laplace distribution
    ZHANG Fagan, HE Youhua
    2021, 27(3):  601-610.  doi:10.12066/j.issn.1007-2861.2125
    Asbtract ( 1441 )   HTML ( 31)   PDF (1247KB) ( 240 )  
    Figures and Tables | References | Related Articles | Metrics

    A new mixed quantile regression model is established using an asymmetric Laplace distribution. Traditional models consider only positional parameters, whereas our model considers the regression of both positional and scale parameters. The expectation maximization (EM) algorithm was used to compute the estimated values of the model parameters. Numerical simulation results showed that the proposed parameter estimation was precise in each quantile, and a larger sample offered higher precision. Our model was applied to the analysis of urban house prices.