Loading...

Table of Content

    Machine Learning & Data Mining
    Video flame detection based on GMM and 3D-LBP feature
    Yunyang YAN,Huizhen ZHANG,Yi′an LIU,Shangbing GAO
    Journal of Shandong University(Engineering Science). 2019, 49(1):  1-9.  doi:10.6040/j.issn.1672-3961.0.2017.430
    Abstract ( 594 )   HTML ( 172 )   PDF (6189KB) ( 282 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to solve the problems of extracting the accuracy of the candidate region and improve the description ability of the flame characteristics, a novel flame detection algorithm based on Gaussian mixture model (GMM) and three-dimensional locality binary pattern (LBP) texture features was proposed. The distribution of flame was analyzed in two spaces of RGB and HSV, and the GMM was trained to extract the flame candidate region. The texture characteristics of the flame was selected as an important feature. The original LBP texture was fused with the motion characteristics of the flame to form a new three-dimensional LBP texture to improve the classification effect of the texture feature on the flame. The one-class support vector machine (One-Class SVM) classification method was used to determine whether the candidate area was a flame.

    Face recognition based on improved prameter-free supervised localitypreserving projections
    Jun FAN,Qiaolin YE,Ning YE
    Journal of Shandong University(Engineering Science). 2019, 49(1):  10-16.  doi:10.6040/j.issn.1672-3961.0.2017.419
    Abstract ( 636 )   HTML ( 7 )   PDF (3094KB) ( 248 )   Save
    Figures and Tables | References | Related Articles | Metrics

    To solve the problem of unsupervised and complexity of parameter selection of the locality preserving projection algorithm, an improved supervised dice parameter-free locality preserving projection algorithm (SdPLPP)was proposed. SdPLPP constructed affinity matrix by using generalized Dice coefficient and extract features of data under the supervised mode, which could avoid the problems of parameters selection and adjustment of locality preserving projection (LPP) algorithm. The proposed algorithm performed experiment of image visualization based on the Iris dataset, analyzed the relationship between the value of the distance of sample data and the performance of the algorithm. To verifying the effectiveness and performance of algorithm, SdPLPP carried out the feature extraction experiments based on three kinds of human face databases, such as ORL, Yale and FERET, and used nearest neighbor classifier to get correct recognition rate. The experimental results showed that the SdPLPP algorithm was superior to PCA, ULDA, LPP, SPLPP and EP-SLPP algorithm in face recognition, and it was better than other algorithms of supervised parameter-free locality preserving projections.

    Improvement of bandwidth model for high speed optical communicationlaser and its optimization by parallel computing
    Si YANG,Sitong LI,Jindong ZHANG,Yu BAI
    Journal of Shandong University(Engineering Science). 2019, 49(1):  17-22, 29.  doi:10.6040/j.issn.1672-3961.0.2018.196
    Abstract ( 641 )   HTML ( 4 )   PDF (3794KB) ( 231 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to adapt to high-speed network, features of optical fiber communication laser need to be optimized. Laser thermal sensing model was improved by measuring the data exported by vertical cavity surface emitting laser (VCSEL). In the modified model, the quantum rate equation was developed to describe features of laser VCSEL. Considering the relationship between the gain constant, the transparent current carriers and the temperature, the high order fitting method was introduced to optimize the laser bandwidth system, which maked it closer to the measured data. Meanwhile, based on the parallel data mining method, the buffer helped to speed up the solving process to satisfy the demand of the next-generation high-speed network. The results showed that the output light power appeared to decrease in thermal saturation under the high temperature and large injection current. The model after optimization was closer to the actual measured data, genetic algorithm had a strong adaptability towards scale change of system, and the route planned by algorithm was reliable. The calculation speed was increased by 9.15% based on parallel data processing method. This modified model met the demand of the new-generation high-speed fiber optic communication, and considered the thermal limitation.

    Detecting frame of repetition forgery based on noise level estimation
    Lala MEI,Ran LI,Chang'an WU
    Journal of Shandong University(Engineering Science). 2019, 49(1):  23-29.  doi:10.6040/j.issn.1672-3961.0.2018.190
    Abstract ( 431 )   HTML ( 13 )   PDF (2297KB) ( 191 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Detecting method of the varying noise level in temporal-domain was investigated based on noise-level, which could identify frame repetition (FR) forgery. Wavelet coefficients were computed for each video frame, and median absolute deviation (MAD) of wavelet coefficients was used to estimate the standard deviation of Gaussian noise mixed in each video frame. Fast Fourier transform (FFT) was used to calculate the amplitude spectrum of the standard deviation curve of the video sequence, and to provide the peak-mean ratio (PMR) of the amplitude spectrum. In order to automatically identify FR forgery, a hard threshold decision based on PMR was taken to determine whether the standard deviation had a periodicity in time domain. The experimental results showed that the proposed method ensured a large PMR for the forged video and high detection accuracy. The proposed method presented a better detection performance when compared with the existing detection, avoiding the performance loss from noise.

    Optimal deployment strategy of forest fire monitoring nodes based on visualization
    Pengcheng ZHAO,Fuquan ZHANG,Xubing YANG,Yin WU
    Journal of Shandong University(Engineering Science). 2019, 49(1):  30-35, 40.  doi:10.6040/j.issn.1672-3961.0.2018.195
    Abstract ( 524 )   HTML ( 5 )   PDF (3136KB) ( 132 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Because of the high cost of forest fire monitoring sensor nodes and large deployment area, the optimization of the deployment efficiency was the main problem in application. In order to coordinate the deployment cost and coverage efficiency of nodes, a visualization-based optimal monitoring nodes deployment strategy was proposed. This strategy was based on a practical dataset from the forest environment, and made viewshed analysis of the candidate nodes. The mutual information algorithm was used to greedily select the location with the highest coverage efficiency by using the viewshed area association matrix of the nodes. The optimal number of nodes was calculated under the cost constraints by the submodular algorithm. This strategy ensured coverage efficiency, and reduced deployment cost. It was a cost-effective deployment strategy for forest fire monitoring nodes.

    Hybrid localization algorithm based on BP neural network and multivariable Taylor series
    Ya'nan YANG,Bin XIA,Nan XIE,Wenhao YUAN
    Journal of Shandong University(Engineering Science). 2019, 49(1):  36-40.  doi:10.6040/j.issn.1672-3961.0.2017.524
    Abstract ( 602 )   HTML ( 4 )   PDF (1585KB) ( 196 )   Save
    Figures and Tables | References | Related Articles | Metrics

    The positioning accuracy of the multivariable Taylor series algorithm depended heavily on the initial values, so a novel hybrid localization algorithm was proposed. The initial values offered by back-bropagation(BP) neural network algorithm could improve the convergence speed of multivariable Taylor series expansion method, and the multivariable Taylor series expansion method could reduce the position error caused by distance measurement error through making full use of the distance information of the unknown nodes. Experimental results indicated that the algorithm could improve positioning accuracy and reduced the influence of mesh spacing on location accuracy.

    Features analysis for Chinese irony detection
    Rongxiang ZHOU,Xiuyi JIA
    Journal of Shandong University(Engineering Science). 2019, 49(1):  41-46.  doi:10.6040/j.issn.1672-3961.0.2018.341
    Abstract ( 679 )   HTML ( 6 )   PDF (1104KB) ( 246 )   Save
    Figures and Tables | References | Related Articles | Metrics

    The research object was data in microblog. The features of irony detection were studied. In view of the characteristics of microblog and irony detection, a variety of features were constructed, such as emotional phrases, emoticons and so on. The experiments showed that the proposed irony features improved 0.34% on recognition accuracy, 0.74% on recall and 0.18% on F-measure, compared with the existing ones for the imbalanced datasets. The proposed irony features also improved 0.44% on recognition accuracy, 2.54% on recall and 0.14% on F-measure, compared with the existing ones for the balanced datasets.

    Advanced collaborative filtering recommendation model based on sentiment analysis of online review
    Chunlin QIAN,Xingfang ZHANG,Lihua SUN
    Journal of Shandong University(Engineering Science). 2019, 49(1):  47-54.  doi:10.6040/j.issn.1672-3961.0.2017.485
    Abstract ( 893 )   HTML ( 19 )   PDF (1086KB) ( 274 )   Save
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the uncertainty of users' subject opinions in online Chinese review, a sentiment analysis model was proposed based on uncertainty theory. An individual recommendation algorithm was designed on the basis of the proposed sentiment analysis model. Firstly, the tokenizers of ICTCLAS and IKAnalyzer were used to preprocess online Chinese review to generate characteristic words, and the point mutual information value of characteristic words accounting for the sentiment direction were computed based on sentiment dictionary (HowNet). Then, the sentiment analysis model was established via uncertainty theory of uncertain variable and uncertain set. In addition, the new similarity formula based on the proposed model was used to search the nearest neighbors. Finally, the recommendation lists were given. The experiments were carried out on two real datasets. The results showed that the proposed method could effectively improve the accuracy of recommendation and alleviate the sparse data problem.

    Control Science & Engineering
    Design for autonomous charging system of family companion robot
    Fengyu ZHOU,Fang WAN,Jiancheng JIAO,Junjian BIAN
    Journal of Shandong University(Engineering Science). 2019, 49(1):  55-65, 74.  doi:10.6040/j.issn.1672-3961.0.2018.301
    Abstract ( 835 )   HTML ( 21 )   PDF (7686KB) ( 374 )   Save
    Figures and Tables | References | Related Articles | Metrics

    To address the limited battery capacity and discontinuous work ring of the family companion robot, a based autonomous charging and docking system was designed for the family companion robot using ROS. The Calman filtering algorithm is used to fuse the encoder data and IMU data. Meanwhile, the environment 2D grid map was constructed by the laser ranger data combined with the Rao-Blackwellized particle filter SLAM algorithm. The global path planning and the local path planning were carried out with the A* approach and the DWA algorithm to control robot reach the neighborhood of the charging station.The dual priority based infrared navigation and docking algorithm was used to guide the robot to the charging station to accurately docking with the charging station. The experimental results showed that the proposed system effectively solved the problem of limited charging distance compared with the traditional method, and had high docking efficiency, success rate, accuracy and generalization ability. Therefore, the system fully satisfied the charge demand of the family companion robot, which could be widely used to address real-world problems.

    Fault detection for unmanned aerial vehicle systems based on strong tracking H-/H optimization
    Xiaoqiang ZHU,Maiying ZHONG
    Journal of Shandong University(Engineering Science). 2019, 49(1):  66-74.  doi:10.6040/j.issn.1672-3961.0.2018.539
    Abstract ( 514 )   HTML ( 5 )   PDF (5445KB) ( 166 )   Save
    Figures and Tables | References | Related Articles | Metrics

    The fault detection method based on the strong tracking H-/H optimization was proposed for a class of unmanned aerial vehicle (UAV) flight control systems. Both the actuator fault and sensor fault were expressed as additive signals, and a nonlinear longitudinal faulty model for a class of UAV was established where the wind disturbance was considered. The design problem of the residual generator was formulated as an extended H-/H optimization one. Inspired by the fast convergence of the strong tracking filter (STF), the STF was introduced to the extended H-/H optimization to realize fast fault detection (FD) for UAV flight control systems. The simulation experiment was carried out with a loss of effectiveness of the elevator and an intermittent fault of the pitot. The simulation results showed that the faults could be detected rapidly through the proposed method. The fault detection method based on the strong tracking H-/H optimization could be applied to the fault detection of UAV flight control systems.

    Optimization of job shop scheduling based on improved particle swarm optimization algorithm
    Hongming LIU,Hongyan ZENG,Wei ZHOU,Tao WANG
    Journal of Shandong University(Engineering Science). 2019, 49(1):  75-82.  doi:10.6040/j.issn.1672-3961.0.2018.540
    Abstract ( 1028 )   HTML ( 28 )   PDF (2579KB) ( 843 )   Save
    Figures and Tables | References | Related Articles | Metrics

    An improved particle swarm optimization algorithm was proposed based on adaptive weights and chaos aiming at the job shop scheduling. A multiple constrained job shop scheduling model was built with the shortest machining time as the optimization goal. The mapping relationship between particle parameters and operation sequences was obtained by coding method based on ranked order value. The inertial coefficient and acceleration factor in the particle swarm optimization algorithm were improved based on the adaptive weights, so that the algorithm could dynamically adjust parameters based on the fitness value. The reverse learning was used to improve the quality of initial solution. Considering the problem of local optimum, some measures were used to enhance the search of algorithm, such as Lévy flight, variable neighborhood search and chaos. The results showed that improved particle swarm optimization could effectively improve utilization of particles, balance the global search and local search, avoid the premature convergence of the traditional particle swarm optimization algorithm, and get better results.

    Civil Engineering
    Study on modification of lake deposition high liquid limit clay in the Yellow River flood area
    Xueyong ZHU,Haiwei LIU,Xiaoyan MA,Qingtao XING,Xiaoqun LIN,Tieqiang MAN
    Journal of Shandong University(Engineering Science). 2019, 49(1):  83-90, 113.  doi:10.6040/j.issn.1672-3961.0.2017.366
    Abstract ( 528 )   HTML ( 3 )   PDF (1461KB) ( 223 )   Save
    Figures and Tables | References | Related Articles | Metrics

    By adding different dosages of lime, fly ash and lime fly ash to the high liquid limit clay in areas formerly flooded by the Yellow River, the liquid-plastic limit test was conducted to investigate the law of material liquid limit and plastic index changing with the number of days, thereby determining the optimal modification dose and effect. The test showed that the main mineral content of high liquid limit soil in areas formerly flooded by the Yellow River was illite, montmorillonite and kaolinite, which had good hydrophily and a liquid limit of about 50%. It had expansibility, easy foaming nature, and poor water stability, with a bearing ratio (CBR) value of less than 3% after 4 d of water immersion; the addition of 4% lime and the 3:9 ratio of lime powder and coal ash in lime-fly ash soil could reduce the hydrophily and expansibility of the material, and improved water stability. The modified soil's CBR reached 8% under the 88% compaction rate after 4 d of water immersion, which met the filling requirements of roadbed area.

    The technology of spanning deep pond of the urban expressway nearby paralleling high-speed railway
    Shulei JIANG,Shifeng YANG,Linghang YANG,Wenming SHI,Fuqing ZOU,Qinghong SHAN
    Journal of Shandong University(Engineering Science). 2019, 49(1):  91-100.  doi:10.6040/j.issn.1672-3961.0.2018.232
    Abstract ( 649 )   HTML ( 2 )   PDF (12682KB) ( 234 )   Save
    Figures and Tables | References | Related Articles | Metrics

    A deep pond section of a city expressway with small interval parallel to the Beijing-Shanghai High-Speed Railway was taken as engineering background. The impact of traditional high embankment on the Beijing-Shanghai High-Speed Railway was analyzed by numerical method from the following four aspects: horizontal displacement and settlement of bridge piers, axial force and skin friction of piles. The results indicated that horizontal displacement of the Beijing-Shanghai High-speed Railway caused by high embankment was far more than the limit, which would cause serious damage to its viaduct. So a spanning technology using low bridges was put forward, which included pre-stressed concrete slab beam, continuous bridge deck erected as simple-supported, gravity abutment, bored pile foundation and 4 m pier. Displacement caused by the low bridge was far less than the limit value. This technology would not affect the normal operation of the Beijing-Shanghai High-speed Railway. Subsequent monitoring results also showed that the low bridge was a reliable technology to spanning deep pond with small interval paralleling to high-speed railway.

    Mechanical Engineering
    Numerical simulation on kinetics characteristics of liquid metal MHD generator
    Yulei ZHANG,Yong WANG,Yudong XIE,Guang SUN,Yanyun WANG,Jiazhen HAN
    Journal of Shandong University(Engineering Science). 2019, 49(1):  101-106.  doi:10.6040/j.issn.1672-3961.0.2018.417
    Abstract ( 681 )   HTML ( 5 )   PDF (1696KB) ( 255 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to get the flowing law in a liquid metal magnetohydrodynamics(MHD) generator channel, three-dimensional numerical studies on kinetics characteristics of open circuit generators were carried out. A magnetic induction method and a modified K-ε turbulence model based on Fluent were adopted to ensure the calculation precision. A comparative analysis of the velocity and force distribution of near-wall and core was made. The velocity fluctuation was defined and different physical parameters were chosen. The velocity profiles and fluctuations downstream of the effective section were analyzed quantitatively. The results showed that, the velocity fluctuations reached maximum at downstream of the effective section and channel parameters had a significant impact on velocity profiles. The channel width being kept constant, the interaction parameter determined the velocity fluctuation and the velocity fluctuation was proportional to the interaction parameter. As channel width increased, the impact of interaction parameters on velocity fluctuations declined.

    Electrical Engineering
    Algorithm of underwater target recognition based on CNN features with BOF
    Wenwen QUAN,Mingxing LIN
    Journal of Shandong University(Engineering Science). 2019, 49(1):  107-113.  doi:10.6040/j.issn.1672-3961.0.2017.385
    Abstract ( 809 )   HTML ( 6 )   PDF (3734KB) ( 306 )   Save
    Figures and Tables | References | Related Articles | Metrics

    In order to prevent false matching problems of scale invariant feature transform (SIFT) matching as a low-level representation for lack of sufficient features, an improved bag of features (BOF) algorithm method combined with the convolution neural network (CNN) features was proposed, which had better semantic segmentation ability to enhance the recognition rates. The LifeCLEF fish video on ImageCLEF website was used to create our own target image databases. Convolution neural network was trained in the Alexnet architecture of caffe, and the features of image databases and query images were extracted. The trained CNN features were simulated in Matlab, and the hamming distance was calculated to verify the matching effect. In addition, the parameter values were changed to test the effect of different Hamming distance thresholds on target recognition results. The experiment of self-made image databases showed that the fusion of depth learning features could effectively improve the underwater target recognition rates of BOF algorithm, and the selection of Hamming distance thresholds required selecting the appropriate parameters according to the actual situation.

    Analysis of insulation state of transformer based on aging factor
    Zhiyong LIN,Damin ZHANG,Xungao ZHONG,Jianbin ZENG,Qiang ZHANG
    Journal of Shandong University(Engineering Science). 2019, 49(1):  114-119.  doi:10.6040/j.issn.1672-3961.0.2018.330
    Abstract ( 500 )   HTML ( 6 )   PDF (1263KB) ( 302 )   Save
    Figures and Tables | References | Related Articles | Metrics

    For further studying effect of extended Debye equivalent circuit on insulation medium aging of transformer, the extended Debye equivalent circuit of transformer was built using the characteristic of return voltage polarization spectrum. By analysis of equivalent circuit parameters' effect on return voltage polarization spectrum, the insulation medium aging of transformer was obtained finally. The simulation and examples analysis results showed that: polarization branch with large time constant reflected the aging of insulation paper in the transformer; polarization branch with small time constant reflected the aging of insulation oil in the transformer. Based on equivalent circuit parameters, aging factor (KL) was defined to analyze insulation state of transformer. The intrinsic relationship between the aging factor and insulation state of transformer was obtained. The analysis results demonstrated that insulation state of transformer was good when KL in range between [0, 10]; Insulation state of transformer was general when KL in range between [10, 30]; Insulation state of transformer was aged seriously when KL in range between [30, ∞].

    Optimization of emergency load shedding of receiving-end power grid based on Particle Swarm Optimization
    Meng LIU,Taoyang XU,Changgang LI,Yue WU,Zhi WANG,Fangfang SHI,Jianjun SU,Guohui ZHANG,Kuan LI
    Journal of Shandong University(Engineering Science). 2019, 49(1):  120-128.  doi:10.6040/j.issn.1672-3961.0.2018.442
    Abstract ( 639 )   HTML ( 5 )   PDF (2721KB) ( 139 )   Save
    Figures and Tables | References | Related Articles | Metrics

    An optimization model and solving method of emergency load shedding were proposed to solve the high voltage direct current (HVDC) blocking of multi-infeed HVDC receiving-end power grid. The model was established to minimize the total amount of shed loads and was constrained by transient frequency variations, voltage variations, line overloading and load capacity of each node to maintain power system stability. The optimal variable of the model was the amount of load shedding of each node. Considering that the model was nonlinear, an adaptive parameter adjustment method based on particle swarm optimization (PSO) algorithm and parallel computing was established. The method improved the global optimization and the speed of the algorithm. Shandong Power Grid model was taken as an example to verify the validity of the algorithm for emergency load shedding following HVDC blocking.