Loading...

Table of Content

      
    20 February 2017
    Volume 47 Issue 1
    A method based on cosine similarity in supplier selection
    ZHANG Li, XIA Peipei, LI Fanzhang
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  1-6.  doi:10.6040/j.issn.1672-3961.1.2016.047
    Abstract ( 1276 )   PDF (1016KB) ( 580 )   Save
    References | Related Articles | Metrics
    An improved TOPSIS(technique for order preference by similarity to ideal solution)based on cosine similarity was proposed and applied to supplier selection. The distance metric based on cosine similarity implicated the weights of attribute in the traditional TOPSIS, and then the proposed method could improve the objectivity of decision. An application on supplier selection was given to show the strong scientificalness and rationality of the proposed method.
    Condition recognition of high-speed train based on multi-view classification ensemble
    GUO Chao, YANG Yan, JIANG Yongquan, SONG Yi
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  7-14.  doi:10.6040/j.issn.1672-3961.1.2016.330
    Abstract ( 1126 )   PDF (2281KB) ( 749 )   Save
    References | Related Articles | Metrics
    To solve the problem about the incompletion of feature extraction and inaccuracy of the identification performance of traditional method, a multi-view classification ensemble method(MVCE)for condition recognition of high speed train was proposed. The method combined with multi-view feature extraction and classification ensemble technology. For condition recognition, the wavelet energy, spectral coefficients and ensemble empirical mode decomposition fuzzy entropy were extracted from three angles: the characteristics of the signal, the frequency domain and the time-frequency domain. The Fisher ratio was used to perform feature selection for the frequency domain features of the high speed train vibration signal, then the feature of the three views were constructed collectively. The least square support vector machine(LSSVM)and the K nearest neighbor(KNN)classifiers were used to recognize each view. The output results of multiple classifiers were integrated by using the classification entropy voting principle. The experimental results showed that the average recognition rate of the proposed method on the simulation data and the laboratory data were 89.18% and 90.87% respectively. Meanwhile, the comparative results illustrated the completeness of the features extracted by the method and the validity of the ensemble model with diversity.
    Random undersampling and POSS method for software defect prediction
    FANG Hao, LI Yun
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  15-21.  doi:10.6040/j.issn.1672-3961.0.2016.304
    Abstract ( 1246 )   PDF (1617KB) ( 610 )   Save
    References | Related Articles | Metrics
    In order to solve the problem of imbalance distribution in software defect prediction, POSS(pareto optimization for subset selection)feature selection and random undersampling was applied in this paper, and SVM was used to build the prediction model. The experimental results showed that the problem could be solved effectively by using multiple random undersampling, and the POSS method was treated subset selection as a bi-objective optimization, which could improve the accuracy of classification, the effectiveness of proposed method was verified by comparing with Relief、Fisher、MI(mutual information).
    A sparse online learning algorithm for feature selection
    WEI Bo, ZHANG Wensheng, LI Yuanxiang, XIA Xuewen, LYU Jingqin
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  22-27.  doi:10.6040/j.issn.1672-3961.1.2016.060
    Abstract ( 1534 )   PDF (2167KB) ( 976 )   Save
    References | Related Articles | Metrics
    In order to effectively deal with mass, high dimensional and sparse big data and improve the efficiency of data classification, an online learning algorithm based on the sparsity principle of L1 norm was proposed. The feature of high dimensional streaming data were novel “Integer” processed by using the online machine learning algorithm framework increased the sparsity of data feature, meanwhile enhanced the partial feature value within the scope of the threshold value and greatly improved the effect of sparse data classification. The performance of SFSOL algorithm was analyzed by using public data sets. The algorithm and the performance of the other three sparse online learning algorithms were compared. The experimental results showed that SFSOL algorithm was more suitable to accurately classify for high-dimensional sparse data.
    Particle network optimization algorithm based on Newtonian mechanics and game theory model
    YI Yunfei, MIAO Jian, LIN Guolong, YIN Zhi
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  28-36.  doi:10.6040/j.issn.1672-3961.1.2016.070
    Abstract ( 1328 )   PDF (2830KB) ( 828 )   Save
    References | Related Articles | Metrics
    In order to overcome the bad convergence and accuracy of the standard particle swarm optimization algorithm in solving the high dimensional TSP problem, each particle was given their own character, such as mass and acceleration was given, and Newton second law with a Poisson distribution was introduced to dynamic control the particle acceleration. In addition, the particle dimensions were divided into advantages and disadvantages section based on its similar to reduce the dimension of update, normally update would only change the disadvantage parts to keep and extend their advantage parts so that it could improve the convergence speed, when the disturbance it would update its advantage parts to away from the current network so that it could jump out of local optimum,when particles collide, opposition-based learning strategy was used to deal with disadvantage section, and a better model of slow convergence was selected. Finally, via numerous simulations of TSPLIB and comparison with other classical algorithms, the results showed that the improved algorithm had the feature of high efficiency, low computational complexity and strong convergence, which were especially crucial for the functioning of large-scale distribution problems. Research results could provide a reference for the study on intelligent algorithms in solving optimization problems, such as how to improve the accuracy and speed up the convergence. 山 东 大 学 学 报 (工 学 版)第47卷 - 第1期易云飞,等:基于牛顿力学和博弈论模型的粒子网络优化算法 \=-
    An improved CNM algorithm based on network structure information
    LYU Zhen, LI Suxue, ZHANG Chuanting, YUAN Dongfeng
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  37-41.  doi:10.6040/j.issn.1672-3961.1.2016.180
    Abstract ( 1778 )   PDF (1231KB) ( 707 )   Save
    References | Related Articles | Metrics
    Although community detection could be effectively accomplished by CNM(clauset-newman-moore)algorithm, the accuracy of the results was unsatisfactory. Consequently, an improved CNM algorithm based on network structure information was proposed, which divided the original network into two parts by removing the edge whose edge betweenness was maximum of all iteratively. These two parts as the input data of CNM algorithm were used to detect communities. The experimental results on five different size of datasets showed that the improved CNM algorithm elevated the quality of community detection, and modularity of these communities peformed well especially in small datasets.
    Android malware detection based on SVM
    ZHANG Yuling, YIN Chuanhuan
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  42-47.  doi:10.6040/j.issn.1672-3961.1.2016.150
    Abstract ( 1161 )   PDF (1961KB) ( 1044 )   Save
    References | Related Articles | Metrics
    In order to detect malware effectively and reduce the threat of malicious software on Android platform security, two strategies that were probability statistics embedding and feature extraction were proposed based on the analysis of existing data sets.These strategies were used to transform high-dimensional data into low-dimensional data so as to reduce the dimension and the uncertainty of the extracted features. Support vector machine were used to classify these data. With these strategies, the time complexity of training process was reduced to 16.7 percent of the original time, and the ability of detecting unknown malware families was improved obviously. Moreover, these strategies were used with some popular classification algorithms, and the experimental results revealed that these strategies could achieve a better detection rate.
    The effects of pile foundation of high-rise building on adjacent structures
    LYU Guoren, ZHANG Qun, NIU Ben, GAO Quanting, WU Zhaoshou
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  48-58.  doi:10.6040/j.issn.1672-3961.0.2016.195
    Abstract ( 1486 )   PDF (4410KB) ( 872 )   Save
    References | Related Articles | Metrics
    In order to study the influence of pile foundation of high-rise buildings on surrounding buildings, a simplified calculation formula for the displacement field calculation of surrounding soil mass of pile foundation was put forth based on existing research theories. The settlement values of a practical ancient building were calculated through the simplified calculation formula to analyze the influence of pile foundation of high-rise building on adjacent structures. The feasibility of the formula was verified by comparing the calculation results of settlement with the practical monitoring results. According to the engineering example, FEM numerical simulation of the incidence of raft and pile length to surrounding buildings was performed. The feasibility and correctness of numerical simulation were also verified by comparing the simulation results to the measured values. The research results provided a reliable basis for the construction of Jinan Shimao Plaza, which could give technical reference for similar geological conditions.
    The control mechanism for super section tunnel on weak surrounding rock and its application
    WAN Li, WANG Chunhe, WANG Qi, LI Shucai, SHAO Xing, JIANG Bei, SUN Huibin, QIN Qian
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  59-67.  doi:10.6040/j.issn.1672-3961.0.2016.232
    Abstract ( 1302 )   PDF (7177KB) ( 599 )   Save
    References | Related Articles | Metrics
    For making clear the failure control mechanism of super section tunnel, numerical experiments of weak rock control mechanism applied to super section tunnel were systematically carried out by the excavation methods that included CRD and double side wall drift methods. The change laws of tunnel surrounding rock deformation and supporting components force were compared and analyzed under four types of supporting schemes and different excavation methods, which included excavation without support, bolt support, H-beam arch support and H-beam-bolt support. The control mechanism of super section tunnel with weak surrounding rock was acquired. Field experiments of H type steel and anchor net shot rete combined support section tunnel were carried out by the excavation methods that including CRD and double side wall drift methods, and the vault settlement were 27.2 mm and 18.7 mm. The results showed that this support system had great effect on the stability control of surrounding rock and could ensure the safety of the site.
    Improvement method of compression test and the dynamic damage of concrete
    LI Ming, ZHU Zhaoquan, LIU Lin
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  68-75.  doi:10.6040/j.issn.1672-3961.0.2016.104
    Abstract ( 1240 )   PDF (2573KB) ( 755 )   Save
    References | Related Articles | Metrics
    In order to explore the incident waveform improvement method in concrete dynamic compression test thus further exploring the dynamic damage and the mechanical property of concrete under the impact loading, the experiment adopted the conventional large diameter split Hopkinson pressure bar(SHPB)to study the dynamic compression test of concrete. The dispersion effect of the incident pulse and the serious high frequency oscillation phenomenon causes large errors in the result of the experiment, but it is the key to ensure the validity of the test and the reliability of the test results to ensure the uniform distribution of stress and to maintain the constant strain rate loading before the test piece destroyed. This study used the brass shaper technology to improve the incident wave to reduce the experimental error and then by controlling variable, then used the method of the combination of theoretical formula and experimental, further researched on the influence of air pressure and impact times on the dynamic damage of concrete. It was concluded that the shaper made of small diameter solid copper thin could improve the accuracy. The damage degree of concrete was gradually augmented with the increase of the repeated impact times the dynamic elastic modulus was decreased, and the ability to resist impact was attenuated.
    The influence of BFRP layers on the bending properties of the reinforced concrete beams
    QIN Zipeng, TIAN Yan, LI Gang, MA Yuwei, LIU Le, ZHANG Jinjian
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  76-83.  doi:10.6040/j.issn.1672-3961.0.2016.186
    Abstract ( 1086 )   PDF (2683KB) ( 586 )   Save
    References | Related Articles | Metrics
    In order to study the impact of different layers of basalt fiber reinforced polymer(BFRP)on the bending performance of reinforced concrete beams, three dividing point loading experiments were performed on the beams with a separate hydraulic jack. The failure modes, crack loads, yield loads, ultimate loads and the mid-span deflections corresponds to the above loads of the beams reinforced with 1, 2 and 3 layers of BFRP at the bottom of the beam were observed respectively and compared with the reference beam. The experiment results showed that, with the increasing number of the pasted BFRP layers, failure modes of the beams were changed, the number of primary cracks were decreased, crack loads, yield loads and ultimate loads were increased with varying degrees. The ultimate loads were respectively increased by 7.92%(1 layer), 23.66%(2 layers)and 29.31%(3 layers), and corresponding span deflections were also respectively decreased by 14.54%, 26.65% and 28.81%. When 1~2 layers of the BFRP were pasted, ductility factors of the beams were increased and the ductility factor was reduced when 3 layers of BFRP were pasted. The generation and expansion of the bending cracks were the main cause of the failure of BFRP. The flexural behavior of the beams was not proportional to the increase of the number of the BFRP layers. Taking together beams strengthened by 2 layers of BFRP, not only the bending bearing capacity and ductility have been greatly improved, but also the deflection has been greatly reduced which has good economic performance.
    Experiment and numerical simulation for surface roughness of 2024 aluminum alloy treated by shot peening
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  84-89.  doi:10.6040/j.issn.1672-3961.0.2016.140
    Abstract ( 1329 )   PDF (3622KB) ( 496 )   Save
    References | Related Articles | Metrics
    In order to research the effects of shot peeing on roughness of 2024 aluminum alloy, an effective method with numerical simulation to predict shot peening roughness was put forward. Aluminum alloy specimens were treated by shot peening, and their roughness Ra were measured by experimental measurement. A model with multiple shots was established to simulate the process of shot peening using ABAQUS, and the Ra were obtained from the statistics of nodal displacements along some sample paths based on the proposed discretized formula. Then the effects of shot parameters on roughness were analyzed through changing the parameters of the model. The results showed that relative error between simulation and experiment were 16.7%, 2.5%, 4.3%, which meant that simulation values were agree well with experimental ones. The correctness of simulation model and results was verified. In addition, it could be concluded by simulation that roughness increased as diameter of shots increased, as velocity of shots increased, and roughness increased quickly at beginning, then increased slowly as shot peening coverage increased.
    The influence of indentation on the surface residual stress of stainless steel
    SU Chenggong, LIU Yan, WANG Weiqiang, WANG Yuhua
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  90-96.  doi:10.6040/j.issn.1672-3961.0.2016.081
    Abstract ( 1259 )   PDF (8386KB) ( 327 )   Save
    References | Related Articles | Metrics
    In order to obtain the residual stress distribution around the residual indentation, the austenitic stainless steel sheets of S30408, S32168, S31603 and S30403 were chosen as the research materials. The constitutive relations of the materials were acquired by the tensile tests, and the force-depth curves were obtained by the ABI tests. The ABI process was simulated by finite element software ABAQUS; The results showed that the maximum value of the residual stress under Mises stress state is about 700 MPa, and the residual stress exists within the area of 1.3% of the ball radius. In addition, the stress concentration around the indentation under the action of external tensile load was studied. The results showed that the maximum superposition stress appears in the indentation materials accumulating area, and the influence range in the thickness direction was limited within 6% of the influence radius.
    Image distortion correction technology of mask image projection stereolithography
    JIA Hongshuai, ZHAO Xuejin, HU Tianliang, ZHANG Chengrui
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  97-103.  doi:10.6040/j.issn.1672-3961.0.2016.313
    Abstract ( 1183 )   PDF (2294KB) ( 640 )   Save
    References | Related Articles | Metrics
    In order to solve image distortion problem when image was projected from image generation device in Mask-image-projection-based stereolithography, a distortion correction method was proposed. Image processing method was used to get pixel position on the projection plane. Predistortion mask image was generated by the intersection operation of 2D contours with pixel actual coordinates. Homogenization of irradiance was achieved by selecting linear interpolation algorithm and bilinear interpolation algorithm according to different pixel position on the projection plane. The experiment results showed that the difference between maximum and minimum irradiance measured value was less than 0.4×100 uW/cm2 and the size error was 0.05 mm, which showed that the designed method was effective and machining accuracy was obviously improved.
    Design of vehicle hydraulic active stabilizer bar and its control algorithm
    KONG Zhenxing, PI Dawei, WANG Xianhui, WANG Hongliang, CHEN Shan
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  104-111.  doi:10.6040/j.issn.1672-3961.0.2016.108
    Abstract ( 1336 )   PDF (3053KB) ( 651 )   Save
    References | Related Articles | Metrics
    According to vehicles active roll control, a design solution to hydraulic active stabilizer bar(ASB)system was put forward based on the analysis of vehicles roll and yaw response characteristics. The structure and principle of hydraulic ASB system was introduced, and its sliding mode control algorithm was designed to improve the vehicles roll stability. The anti-roll torque of ASB system on the front and rear axles were distributed dynamically to improve vehicles steering characteristics. Based on MATLAB/Simulink, 14 degree-of-freedom vehicle, hydraulic actuators model, road input model and so on were established. The simulation of PID+feedforward control and sliding mode control system was carried out under typical maneuvers. Simulation results showed that the hydraulic ASB system using sliding mode control algorithm had obvious advantage in the robustness and adaptability compared with PID+feedforward control, which could enhance vehicle roll and yaw response effectively, and improve roll stability, ride comfort and handling stability further.
    Payload pendulation dynamic analysis of ship-mounted crane
    HAN Guangdong, ZHANG Tong, CHEN Haiquan, WANG Shenghai, ZHANG Jinnan
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  112-118.  doi:10.6040/j.issn.1672-3961.0.2016.326
    Abstract ( 1442 )   PDF (6642KB) ( 723 )   Save
    References | Related Articles | Metrics
    The pendulation of payload was caused by the motion of ship which induced by wave. The motion equation was obtained by means of the relative position of the payload, The kinematics model of ship-mounted crane payload system was established using Lagrange equation, the dynamic was carried out in Matlab/Simulink simulation environment, the influence of rope length, excitation frequency and lifting speed on payload pendulation was analyzed in a comparative manner. The simulation results showed that the payload swing angle increased first and then decreased with the increasing of the rope length; meanwhile, the swing angle increased with the excitation frequency approaching the natural frequency of the system; finally, the swing angle tended to increase with the increasing of the lifting speed, however, the swing angle decreased with the increasing of the lowering speed. Experiment was carried out, and the accuracy of the simulation results was verified.
    Design and development of test bench for hybrid electric vehicle powertrain
    CHEN Yucheng, SUN Qiang, MIAO Qiang, BAI Shuzhan, LI Guoxiang
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  119-124.  doi:10.6040/j.issn.1672-3961.0.2016.121
    Abstract ( 1347 )   PDF (2033KB) ( 588 )   Save
    References | Related Articles | Metrics
    In order to develop hybrid energy management, a single-axis parallel hybrid powertrain test bench based on the idea of modularization was developed. According to the analysis of vehicle driving cycle, major power components were selected and their basic parameters were defined. The resistance simulation module was designed based on the vehicle resistance formula and analysis of transmission system. The control network, composed of vehicle control unit and power unit controllers, communicated with other parts through CAN bus. The results showed that the actual speed followed the target one well. Data acquisition frequency and error conformed to design requirements. The test bench could be used to realize the simulation vehicle driving cycle and test transient conditions.
    Study on matching property of supports and stiffeneron circular exhaust duct
    QI Jinsheng, AN Chunguo, BAI Jie, WANG Zhan
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2017, 47(1):  125-130.  doi:10.6040/j.issn.1672-3961.0.2016.264
    Abstract ( 1239 )   PDF (4906KB) ( 660 )   Save
    References | Related Articles | Metrics
    The matching property of saddle support and stiffener on circular exhaust duct was studied, and the optimal position of heavy stiffener and saddle supports were determined by the finite element method(FEM). The results showed that the use of pure saddle supports might lead to stress concentration on the contact area of saddle and duct, meanwhile the maximum stress could exceed the allowable stress. The heavy stiffener near the supports could decrease the deformation and increase the buckling factor, while could hardly decrease the maximum stress. Saddle supports under the heavy stiffener could obviously decrease the deformation and increase the buckling factor, and could decrease the stress on the contact area of support and duct so that would meet the need of long-term security. The influence of wrap-angle to the stress concentration and buckling property was not obvious.