Table of Content

    28 May 2021, Volume 55 Issue 5 Previous Issue    Next Issue
    Wave-Induced Seabed Response and Liquefaction Around Pipeline at Different Buried Depths
    ZHANG Qi, ZHOU Xianglian, YE Guanlin
    2021, 55 (5):  489-496.  doi: 10.16183/j.cnki.jsjtu.2019.349
    Abstract ( 801 )   HTML ( 868 )   PDF (1126KB) ( 506 )   Save

    In order to study the liquefaction of the seabed around the pipeline under wave loading, a two-dimensional numerical model was established based on Biot’s partly dynamic equation (u-p model). The dynamic response of the seabed around the pipeline under wave loading was investigated in detail, and the wave loading was applied on the seabed surface through pore pressure boundary. Based on the validation of the numerical model, the response and liquefaction of the seabed around the pipeline under wave loading were studied. The pore pressure, vertical effect stress, and liquefaction range of the seabed at different pipeline buried depths were investigated and the effects of wave height, soil permeability, and saturation were discussed. The results show that the pipeline buried depth has significant effects on the response and liquefaction of seabed under wave loading. The pipeline induces the vertical effective stress concentration of seabed around the pipeline. The effects of wave height, soil permeability, and saturation on the seabed response under wave loading are significant. The results provide a theoretical basis for the safety and stability of submarine pipeline in marine environment.

    Figures and Tables | References | Related Articles | Metrics
    Comparative Test of Characteristics of Vortex-Induced Motion and Galloping of Classic Spar Platform
    ZHANG Chenya, KOU Yufeng, LÜ Haining, XIAO Longfei, LIU Mingyue
    2021, 55 (5):  497-504.  doi: 10.16183/j.cnki.jsjtu.2020.206
    Abstract ( 727 )   HTML ( 11 )   PDF (2897KB) ( 388 )   Save

    In order to study the characteristics of vortex-induced motion and galloping of the classic Spar platform, a model test with a mooring system is conducted in basin. The sway motion characteristics of the Spar platform at different current velocities are analyzed. By comparing the results in currents, waves, and wave-current coupled conditions, the coupling effects of current and wave on vortex-induced motion and galloping of the Spar platform are also studied. The results show that galloping is induced by the currents with high reduced velocities. Compared with the vortex-induced motion, the galloping phenomenon has a longer period, larger amplitude, and randomness. The coupling of current and wave would not change the mode of flow-induced motion, but it significantly affects the motion amplitude.

    Figures and Tables | References | Related Articles | Metrics
    Ice Floe Trajectory Under the Action of Propeller Pumping
    WANG Chao, YANG Bo, WANG Chunhui, GUO Chunyu, XU Pei
    2021, 55 (5):  505-512.  doi: 10.16183/j.cnki.jsjtu.2019.365
    Abstract ( 627 )   HTML ( 9 )   PDF (8916KB) ( 303 )   Save

    In order to study the influence of propeller pumping action on ice trajectory during the process of ice propeller interference, this paper built a motion trajectory measurement platform based on circulating water tank, combining high-speed camera and the Photron FASTCAM Analysis (PFA) method, and tested and analyzed the trajectory of ice at different propeller rotation speeds. After systematic analysis of the test results, it is found that when the propeller rotates at a high speed and the flow velocity is small, the pumping action effect is obvious, which apparently changes the model ice trajectory and even makes the ice and propeller collide. When the model ice volume is large and the water flow velocity is fast, the propeller pumping action effect is small, and it is difficult for the pumping action effect to change the ice trajectory because of the influence of the water flow velocity, which has little effect on the model ice motion trajectory.

    Figures and Tables | References | Related Articles | Metrics
    Simulation of Undercooling Influence on Shape and Velocity of Seawater Freezing Process Using Phase Field Method
    BAI Xu, YANG Sujie
    2021, 55 (5):  513-520.  doi: 10.16183/j.cnki.jsjtu.2020.077
    Abstract ( 768 )   HTML ( 11 )   PDF (4210KB) ( 501 )   Save

    In order to study the microcosmic mechanism of ice formation in ship structure, a numerical simulation is conducted. Based on the Wheeler phase field model, the finite difference method is used to simulate the solidification shape of seawater at different undercoolings. In the process of simulation, sea water is regarded as the binary mixture of salt and pure water, and the authenticity of ice crystal growth is ensured by setting ice physical parameters and introducing crystal nucleus. In the calculation, four nuclei are set up to simulate the ice crystal growth, and the influence of different undercoolings on the ice crystal growth is discussed. The results show that the growth rate of ice crystal increases with the increase of undercooling, and it is linear with the change of undercooling. When the dimensionless undercooling is less than 0.85, the dendrite has only a few branches which are thin. When the dimensionless undercooling is higher than 0.85, the ice crystal has more than two-stages of branches which are obviously thicker. When the dimensionless undercooling reaches 1.0, the branches are squeezed by the main branches and lessned, and the crystal cores are finally filled.

    Figures and Tables | References | Related Articles | Metrics
    Numerical Investigation of Three-Dimensional Shallow-Water Sloshing Based on High Accuracy Boussinesq Equations
    YUAN Xinyi, SU Yan, LIU Zuyuan
    2021, 55 (5):  521-526.  doi: 10.16183/j.cnki.jsjtu.2020.053
    Abstract ( 615 )   HTML ( 9 )   PDF (1430KB) ( 514 )   Save

    Highly accurate Boussinesq-type equations in terms of velocity potential are used for the simulation of shallow-water sloshing in a three-dimensional tank under the framework of the potential flow theory. The total velocity potential is separated into two parts: one part is a particular solution which satisfies the Laplace equation in the fluid domain and the no-flow condition on the walls while the other part is solved by the Boussinesq-type model. In the process of numerical calculation, the finite difference method is used for spatial derivative discretization and the 4th Runge-kutta method is used for time iteration. To verify the numerical model, the aspect ratio of the tank is set to be much less than 1 for simulation of 2D cases and is compared with the results published. In the 3D cases, four different sloshing motion forms are observed at each external excitation frequency, and a corresponding number of traveling waves are observed on the free surface. Moreover, the effects of external excitation frequency and coupling excitation on the sloshing motion in the tank are discussed.

    Figures and Tables | References | Related Articles | Metrics
    Method for Plate Crack Damage Detection Based on Long Short-Term Memory Neural Network
    ZHANG Songlin, MA Dongliang, WANG Deyu
    2021, 55 (5):  527-535.  doi: 10.16183/j.cnki.jsjtu.2020.095
    Abstract ( 648 )   HTML ( 18 )   PDF (1411KB) ( 321 )   Save

    Aimed at the problem of intelligent classification of crack damage in different positions of the plate, a method for plate crack damage detection based on long short-term memory (LSTM) neural network is proposed. The Abaqus secondary development is used to build the plate crack damage model and calculate the acceleration response of the plate under Gaussian white noise excitation. The data set is generated by data augmentation, and the influence of noise on damage detection is considered. An intelligent crack detection model based on LSTM is established, which directly takes the acceleration response of the plate as the input and does not require additional damage feature extraction. With the goal of minimizing prediction error, the hyperparameter of the model is selected and the model configuration is optimized. The comparison of the multi-layer perceptron model and the multi-layer perceptron model based on wavelet packet transform shows that the LSTM model proposed in this paper has a higher damage location accuracy and a better applicability in plate crack detection.

    Figures and Tables | References | Related Articles | Metrics
    Promotion of a No Fit Polygon Algorithm Based on Trajectory
    ZHOU Xin, LAI Xiaoyang, MENG Xiangqun, WANG Kun, TANG Houjun
    2021, 55 (5):  536-543.  doi: 10.16183/j.cnki.jsjtu.2020.208
    Abstract ( 1034 )   HTML ( 28 )   PDF (1564KB) ( 543 )   Save

    The nesting problem is how to nest objects of a specified shape in a two-dimensional space to obtain the maximum space utilization rate, which is of great significance in industrial production. The solution to the nesting problem requires frequent cross-checking of the objects to determine whether the nesting position is legal. The No Fit Polygon algorithm can be used to accelerate the procedure of cross-checking, but the algorithm cannot be used to calculate shapes containing curves, which limits its application. An algorithm based on orbit sliding can calculate No Fit Polygon between shapes which includes arc, but its effectiveness is not satisfactory. Aimed at this problem, and based on trajectory algorithm, the trajectory generation strategy and the profile algorithm are analyzed and improved. The improved algorithm can calculate the No Fit Polygon between shapes containing arc in a shorter time, solving both accuracy and effectiveness problems. Finally, the algorithm is tested in the real punch production process and the results of the test confirms the correctness and effectiveness of the algorithm.

    Figures and Tables | References | Related Articles | Metrics
    Dynamic Knowledge Graph Modeling Method for Ship Block Manufacturing Process
    SONG Dengqiang, ZHOU Bin, SHEN Xingwang, BAO Jinsong, ZHOU Yaqin
    2021, 55 (5):  544-556.  doi: 10.16183/j.cnki.jsjtu.2020.241
    Abstract ( 920 )   HTML ( 28 )   PDF (6895KB) ( 870 )   Save

    In the dynamic and discrete ship block manufacturing process, lack of effective process resource organization and transparency in product processing leads to the problem of high cost and low efficiency for managers to acquire knowledge. A method for dynamic generation and updating of knowledge graph based on processing beat data flow is proposed. The definition of the processing beat data information model is defined by analyzing the processing flow and the station data characteristics of the ship blocks. The graph mapping steps, models, and fusion connection algorithms are proposed for static resources and processing beat data to realize the semantic association of station dynamic time series data and knowledge graphs. Based on the relationship between station process and product structure, the generation of workshop-level dynamic knowledge graph is realized. Taking the production process of a ship block as an example, the knowledge graph visualization prototype system is designed, developed, and verified. The results show that the proposed method is beneficial to the organization, acquisition, and reuse of knowledge in the process of ship block manufacturing.

    Figures and Tables | References | Related Articles | Metrics
    Unbalanced Learning of Generative Adversarial Network Based on Latent Posterior
    HE Xinlin, QI Zongfeng, LI Jianxun
    2021, 55 (5):  557-565.  doi: 10.16183/j.cnki.jsjtu.2019.264
    Abstract ( 703 )   HTML ( 11 )   PDF (1888KB) ( 393 )   Save

    Based on the problem that the oversampling method in the existing unbalanced classification problem cannot fully utilize the data probability density distribution, a method named latent posterior based generative adversarial network for oversampling (LGOS) was proposed. This method used variational auto-encoder to obtain the approximate posterior distribution of latent variable and generation network could effectively estimate the true probability distribution function of the data. The sampling in the latent space could overcome the randomness of generative adversarial network. The marginal distribution adaptive loss and the conditional distribution adaptive loss were introduced to improve the quality of generated data. Besides, the generated samples as source domain samples were put into the transfer learning framework, the classification algorithm of transfer learning for boosting with weight scaling (TrWSBoost) was proposed, and the weight scaling factor was introduced, which effectively solved the problem that the weight of source domain samples converge too fast and lead to insufficient learning. The experimental results show that the proposed method is superior to the existing oversampling method in the performance of common metrics.

    Figures and Tables | References | Related Articles | Metrics
    Modeling and Experimental Study on Static Characteristics of Bionic Pneumatic Muscle Fiber
    LEI Jingtao, ZHANG Yuewen, DAI Zhenhao, XU Zili
    2021, 55 (5):  566-574.  doi: 10.16183/j.cnki.jsjtu.2020.092
    Abstract ( 726 )   HTML ( 8 )   PDF (5142KB) ( 559 )   Save

    In this paper, the static characteristics modeling and experimental study of pneumatic muscle fiber (PMF) are conducted. Considering the influence of end deformation, friction, and dead zone pressure on the static characteristics of PMF, a mathematical model of static characteristics of PMF is proposed. The static characteristics experimental platform is designed, and the static characteristics experiments of PMF and pneumatic muscle fiber bundles (PMFB) are performed, including the isometric experiment, the isotonic experiment, and the isobaric experiment. The static characteristics of PMF and PMFB with different specifications are compared and analyzed. Based on the isobaric characteristic curves obtained from the experiment, an experimental model of PMFB is proposed. Besides, the mathematical models of static characteristics of PMF and PMFB are identified by a large number of experimental data, which are in line with the actual situation. The study in this paper will lay the foundation for the precise control of micro bionic robot driven by PMF.

    Figures and Tables | References | Related Articles | Metrics
    A Regionalization Vision Navigation Method Based on Deep Reinforcement Learning
    LI Peng, RUAN Xiaogang, ZHU Xiaoqing, CHAI Jie, REN Dingqi, LIU Pengfei
    2021, 55 (5):  575-585.  doi: 10.16183/j.cnki.jsjtu.2019.277
    Abstract ( 764 )   HTML ( 16 )   PDF (4210KB) ( 593 )   Save

    Aimed at the problems of navigation in distributed environment of a mobile robot, a regionalization vision navigation method based on deep reinforcement learning is proposed. First, considering the characteristics of distributed environment, the independent submodule learning control strategy is used in different regions and the regionalization model is built to switch and combine navigation control strategies. Then, in order to make the robot have a better goal-oriented behavior, reward prediction task is integrated into the submodule, and reward sequence is played back in combination with the experience pool. Finally, depth limitation is added to the primitive exploration strategy to prevent the traversal stagnation caused by collision. The results show that the application of reward prediction and depth obstacle avoidance is helpful to improve navigation performance. In the process of multi-area environment test, the regionalization model shows the advantages that the single model does not have in terms of training time and rewards, indicating that it can better deal with large-scale navigation. In addition, the experiment is conducted in the first-person 3D environment, and the state is partially observable, which is conducive to practical application.

    Figures and Tables | References | Related Articles | Metrics
    A Heterogeneous Network Representation Method Based on Variational Inference and Meta-Path Decomposition
    YUAN Ming, LIU Qun, SUN Haichao, TAN Hongsheng
    2021, 55 (5):  586-597.  doi: 10.16183/j.cnki.jsjtu.2020.187
    Abstract ( 720 )   HTML ( 10 )   PDF (3660KB) ( 329 )   Save

    Aimed at the problem that the traditional meta-path random walk in heterogeneous network representation cannot accurately describe the heterogeneous network structure and cannot capture the true distribution of network nodes well, a heterogeneous network representation method based on variational inference and meta-path decomposition is proposed, which is named HetVAE. First, combining with the idea of path similarity, a node selection strategy is designed to improve the random walk of the meta-path. Next, the variational theory is introduced to effectively sample the latent variables in the original distribution. After that, a personalized attention machanism is implemented, which weights the node vector representation of different sub-networks obtained by decomposition. Then, these node vectors are fused by the proposed model, so that the final node vector representation can have richer semantic information. Finally, several experiments on different network tasks are performed on the three real data sets of DBLP, AMiner, and Yelp. The effectiveness of the model is verified by these results. In node classification and node clustering tasks, compared with some state-of-the-art algorithms, the Micro-F1 and normalized mutual information (NMI) increase by 1.12% to 4.36% and 1.35% to 18% respectively. It is proved that HetVAE can effectively capture the heterogeneous network structure and learn the node vetcor representation that conforms more with the true distribution.

    Figures and Tables | References | Related Articles | Metrics
    Collective Data Anomaly Detection Based on Reverse k-Nearest Neighbor Filtering
    WU Jin’e, WANG Ruoyu, DUAN Qianqian, LI Guoqiang, JÜ Changjiang
    2021, 55 (5):  598-606.  doi: 10.16183/j.cnki.jsjtu.2020.011
    Abstract ( 770 )   HTML ( 5 )   PDF (1494KB) ( 443 )   Save

    Aimed at the problem of group data anomaly detection with no data labels, a k-nearest neighbor (kNN) algorithm is proposed to detect group data anomalies in the unsupervised mode. In order to reduce false negatives and false positives caused by the mutual interference between abnormal and normal values, a reverse k-nearest neighbor (RkNN) method is proposed to filter the abnormal group data in reverse. First, the RkNN algorithm uses statistical distance as the similarity measure between different groups of data. Then, the anomaly scores of each group and the initial abnormality are obtained by using the kNN algorithm. Finally, the initial abnormality is filtered by using the RkNN method. The experiment results show that the algorithm proposed can not only effectively reduce the false negatives and false positives, but also has a high anomaly detection rate and good stability.

    Figures and Tables | References | Related Articles | Metrics
    Video Abnormal Detection Combining FCN with LSTM
    WU Guangli, GUO Zhenzhou, LI Leiting, WANG Chengxiang
    2021, 55 (5):  607-614.  doi: 10.16183/j.cnki.jsjtu.2020.120
    Abstract ( 955 )   HTML ( 15 )   PDF (6753KB) ( 526 )   Save

    In view of the shortcomings of the traditional video anomaly detection model, a network structure combining the fully convolutional neural (FCN) network and the long short-term memory (LSTM)network is proposed. The network can perform pixel-level prediction and can accurately locate abnormal areas. The network first uses the convolutional neural network to extract image features of different depths in video frames. Then, different image features are input to memory network to analyze semantic information on time series. Image features and semantic information are fused through residual structure. At the same time, the skip structure is used to integrate the fusion features in multi-mode and upsampling is conducted to obtain a prediction image with the same size as the original video frame. The proposed model is tested on the ped 2 subset of University of California, San Diego (UCSD) anomaly detection dataset and University of Minnesota System(UMN)crowd activity dataset. And both two datasets achieve good results. On the UCSD dataset, the equal error rate is as low as 6.6%, the area under curve reaches 98.2%, and the F1 score reaches 94.96%. On the UMN dataset, the equal error rate is as low as 7.1%, the area under curve reaches 93.7%, and the F1 score reaches 94.46%.

    Figures and Tables | References | Related Articles | Metrics
    Multi-Scale Reliability-Based Design Optimization of Three-Dimensional Orthogonal Woven Composite Fender
    TAO Wei, LIU Zhao, XU Can, ZHU Ping
    2021, 55 (5):  615-623.  doi: 10.16183/j.cnki.jsjtu.2019.283
    Abstract ( 643 )   HTML ( 8 )   PDF (7652KB) ( 352 )   Save

    Three-dimensional orthogonal woven composites have excellent mechanical properties and delamination resistance, which have a bright future in the application of automotive lightweight. A prediction model of elastic properties for three-dimensional orthogonal woven composites was established based on the analytical micromechanical method. The macro-scale performances of the fender were analyzed by using the finite element method. The Monte Carlo reliability analysis method, the Kriging surrogate mode, and the particle swarm optimization algorithm were adopted to conduct multi-scale reliability-based design optimization of a composite structure, which involves the uncertainties of material and structural design variables. The results show that the optimized fender meets the requirments of structural stiffness and reliability, and it also achieves a weight reduction of 21.93%.

    Figures and Tables | References | Related Articles | Metrics
    Innovation and Design
    Development of a Triboelectric Performance Test System
    HE Ke, WU Zishuai, WANG Daoai, ZHANG Zhinan
    2021, 55 (5):  624-630.  doi: 10.16183/j.cnki.jsjtu.2020.399
    Abstract ( 683 )   HTML ( 4 )   PDF (11741KB) ( 424 )   Save

    In view of the lack of standardized test support for existing triboelectric research, a system suitable for the triboelectric performance test was independently developed. Through modular design and system integration, the three modules of loading, motion, and measurement control were designed in turn. The LabVIEW measurement control software was developed. By adding buffer springs and optimizing the loading structure, the stable loading under small load conditions was achieved. In addition, based on the insulation treatment of the upper and lower samples, the accurate measurement of micro currents was achieved. The reliability of the test system was verified by benchmarking test and friction electrification test, and the transferred charge of Cu-Al friction interface was analyzed. Moreover, the linear relationship between transferred charge and load was preliminarily determined. The results are beneficial to the standardization of tribo-electrical testing.

    Figures and Tables | References | Related Articles | Metrics