Table of Content

  

20 April 2016
Volume 46 Issue 2

Radial basis function extreme learning machine based on core sets

ZHAI Junhai, ZHANG Sufang, HU Wenxiang, WANG Xizhao

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  1-5.  doi:10.6040/j.issn.1672-3961.2.2015.030

Abstract ( 1716 )   HTML   PDF (2001KB) ( 1319 )   Save

References | Related Articles | Metrics

Radial basis function-extreme learning machine(RBF-ELM)employed randomized method to initialize the centers and widths. Randomly initialization of the two parameters led to instability of RBF-ELM. Moreover, for different data sets, it was difficult to determine the number of the hidden nodes. An improved algorithm was proposed, which firstly selected important instances with core set method, and then the centers were initialized with the selected instances, the width parameters were randomly initialized. The proposed algorithm not only could solve the problem of the instability of RBF-ELM to some extent, but also could determine the number of hidden layer nodes. Experimental results showed that the proposed algorithm outperforms RBF-ELM algorithm.

Face image quality evaluation method based on the fusion of two level evaluation indexes

ZOU Guofeng, FU Guixia, LI Zhenmei, LI Haitao, WANG Kejun

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  6-13.  doi:10.6040/j.issn.1672-3961.1.2015.160

Abstract ( 1842 )   HTML   PDF (1846KB) ( 1617 )   Save

References | Related Articles | Metrics

To overcome the influence of face pose and illumination, the face symmetry degree evaluation method based on sub-regions histogram distance was proposed, which could be used to evaluate the effect on face quality of asymmetric illumination and pose. A novel evaluation strategy was proposed, which was the fusion of the first level evaluation index for the original natural image including face and the second level evaluation index for face effective area. The feedback information of the first level evaluation could effectively guide the construction and improvement of image acquisition environment, to provide high quality image source for the latter part of face detection and recognition. The main evaluation indexes included contrast degree, suitability degree, symmetry degree, clarity and effective area. Experiment results showed that the proposed symmetry degree evaluation method and the fusion strategy of two level evaluation indexes had good reliability and accuracy.

Object classification method based on component pyramid matching

ZHU Jie, WANG Jing, LIU Fei, GAO Guandong, DUAN Qing

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  14-21.  doi:10.6040/j.issn.1672-3961.2.2015.065

Abstract ( 1190 )   HTML   PDF (5026KB) ( 639 )   Save

References | Related Articles | Metrics

The image representation method based on component pyramid matching(CPM)was proposed, which separated the patches into different levels based on colors. In each level, some colors were selected by the optimal color selection method, then the patches with these selected colors were considered as the foreground components, and the rest of the patches with other colors were considered as the background components. Usually, the foreground components corresponded to some parts of the objects, which could supply weak semantic information for the image representation. Then, the background components were split into the foreground and background components in the next level based on the similar color selection method. The final representation of an image was obtained by concatenating the component histograms in each level. Classification results were presented on Soccer, Flower17 and Flower102 datasets, and the experiments showed that CMP could obtain satisfactory results in these datasets.

Design patterns detection based on FCA and CBR

XIAO Zhuoyu, HE Pei, YU Bo, LI Yan, HU Zhentao

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  22-28.  doi:10.6040/j.issn.1672-3961.1.2015.046

Abstract ( 1527 )   HTML   PDF (1373KB) ( 1482 )   Save

References | Related Articles | Metrics

Aiming to obtain the accurate and reliable detecting model of design patterns that fusion formal concept analysis(FCA)techniques and case cased reasoning(CBR), a novel refinement technique based on more complete software problem description was proposed. Indexes and cases similarity score value was calculated by FCA and Cosine theory. The results of the priority achieved optimal choices, the new knowledge for the retention process phase of the learning model was provided. An approach based on mean average precision(MAP)to assess the performance was proposed. Finally, the experimental results showed that the presented model had more detecting ability in term of MAP comparing to the traditional models.

A weight-based initial centers selection algorithm for K-modes clustering

JIANG Feng, DU Junwei, LIU Guozhu, SUI Yuefei

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  29-34.  doi:10.6040/j.issn.1672-3961.0.2015.101

Abstract ( 1575 )   HTML   PDF (886KB) ( 951 )   Save

References | Related Articles | Metrics

The current initialization methods for K-modes clustering do not consider the case in which various attributes have different significances. To solve this problem, a weighted density and weighted overlap distance-based initial center selection algorithm(called Ini_-Weight)was proposed. In algorithm Ini_-Weight, initial centers were selected by calculating the density of each object and the distance between any two objects. In Ini_-Weight, when calculating the density of each object and the distance between any two objects, different weights were assigned to different attributes according to the significance of each attribute. Finally, Ini_-Weight was compared with the current methods on UCI data sets. The results showed that Ini_-Weight algorithm could effectively distinguish different attributes and improve the accuracy for selecting initial centers.

A method of opinion leaders discovering based on the topical evolution

WANG Yijun, ZHANG Hui, LI Bo, YANG Chunming, ZHAO Xujian

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  35-42.  doi:10.6040/j.issn.1672-3961.1.2015.165

Abstract ( 1619 )   HTML   PDF (1755KB) ( 1125 )   Save

References | Related Articles | Metrics

Existing studies about opinion leaders mining only adopted complex network methods or built graph model from networks, which ignored the specific role that opinion leaders play in the evolution of public opinion. To solve this problem, an opinion leaders discovering method based on the topic all evolution was presented. First, this method found latent opinion leaders by building a graph model based on the interaction between users and using the graph theory algorithm of looking for central node. Second, the evolution model of public opinion was used to judge the opinion representation of these opinion leaders. Finally, the true opinion leaders were found out, which had real lead resistance in the overall evaluation of public opinion. Experiment based on Sina MicroBlog datasets showed that this method performed better than those methods only consider the network model.

Entropy-based collaborative filtering algorithm

ZHANG Jia, LIN Yaojin, LIN Menglei, LIU Jinghua, LI Huizong

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  43-50.  doi:10.6040/j.issn.1672-3961.2.2015.047

Abstract ( 1630 )   HTML   PDF (1368KB) ( 1836 )   Save

References | Related Articles | Metrics

In the recommender system, the recommended quality was restricted by the sparsity of user rating data. To solve this problem, a novel entropy-based collaborative filtering algorithm was proposed. First, the definition of user entropy was given to reflect the rating distribution of users and their rating tendency degree. Then, the method of large margin was introduced to calculate the margin distance, and the neighbor selection range was determined via combining both of the active users entropy and margin distance with other users. Finally, neighbors were obtained by making full of the user entropy and the similarity between users, which could degrade the influence of the sparse rating data. Experimental results on two data sets showed that the proposed algorithm could improve the recommended quality effectively.

Mulit-Agent dynamic task allocation based on improved contract net protocol

LI Ming, LIU Wei, ZHANG Yanduo

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  51-56.  doi:10.6040/j.issn.1672-3961.1.2015.218

Abstract ( 1519 )   HTML   PDF (1618KB) ( 1373 )   Save

References | Related Articles | Metrics

To apply the multi-agent system task allocation algorithm to dynamic environment, an improved contract net protocol was proposed. First, the agent capability model and the tasks description of agent execution were studied. Secondly, the agent changed its capability dynamically by putting the executing tasks out to tender, so that the tasks could be redistributed. Finally, the feasibility and effectiveness of the proposed method were proposed by building a simulation system of AGV warehouse. The results of simulation indicated that the proposed method could reduce the total time for fulfilling the tasks effectively, improve the overall efficiency of the system and improve the capability of multi-agent system to adapt to dynamic environment.

An endpoint detection algorithm based on frequency-domain characteristics and transition fragment judgment

GUO Yu, ZHANG Erhua, LIU Chi

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  57-63.  doi:10.6040/j.issn.1672-3961.2.2015.147

Abstract ( 1449 )   HTML   PDF (2077KB) ( 1124 )   Save

References | Related Articles | Metrics

In order to improve the accuracy of speech endpoint detection as well as enhance robustness of the endpoint detection algorithm in noisy environment, two new endpoint detection parameters were proposed. The spectrum entropy based on critical band took both perceptual characteristics of the human auditory system and the differences between speech and noise signals in frequency domain distribution into account, as well as the minus frequency-domain energy parameter paid attention to the difference between speech frames and silence frames in frequency energy. The advantages of those two parameters were combined to constitute a robust endpoint detection parameter. Meanwhile, in order to avoid the miscarriage of judgment caused by the unitary threshold, the transition fragment judgment based on statistics of characteristics distribution was applied. The experiment results showed that the endpoint detection algorithm had better discrimination for speech frames and silence frames, the algorithm could carry out better accuracy than other conventional anti-noisy endpoint detection algorithms under different and low signal-to-noise ratio noisy environments, especially in the case of non-stationary noise, the accuracy improved by more than 5%.

Multi-machine power system excitation control based on continuous higher-order sliding mode

LIU Xiangjie, HAN Yaozhen

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  64-71.  doi:10.6040/j.issn.1672-3961.0.2016.011

Abstract ( 1625 )   HTML   PDF (1681KB) ( 879 )   Save

References | Related Articles | Metrics

A continuous higher-order sliding mode excitation control strategy was proposed for improving transient stability of multi-machine power system. Each power angle deviation was chosen as sliding variable. Higher-order sliding mode control for nonlinear and uncertain multi-machine power system was converted to finite time stability problem of uncertain integral chain. The controller was composed of geometric homogeneous continuous control law and second-order sliding mode super-twisting algorithm to achieve finite time convergence and conquer system uncertainties. The derivatives of power angles were estimated by so-called exact robust differentiator. Finite time stabilization of closed-loop system was theoretically proved. The proposed excitation control scheme could stabilize terminal voltage and enhance transient stability effectively. Simulation results for a three-machine system verified the validity of the proposed control method.

An interval UTA method based on the satisfaction degree of decision maker

XIONG Wentao, FENG Yuqiang

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  72-77.  doi:10.6040/j.issn.1672-3961.0.2015.381

Abstract ( 1569 )   HTML   PDF (880KB) ( 904 )   Save

References | Related Articles | Metrics

An interval UTA method was proposed for inferring interval utility functions from a partial preorder of alternatives evaluated on multiple criteria, which was an extension of the well-known UTA method capable to handle the interval evaluation data. Firstly, according to the original UTA method, the interval attribute values of all reference options were transformed into the ranges of utility, namely, the utility intervals. Next, the overall utility intervals were calculated using the arithmetic operations of interval number. A linear programming model was constructed based on the satisfaction degree of decision maker utilizing the mid-point and half-width of interval numbers. After the total error value was obtained, a quadratic programming model was established in the post-optimization step, where the objective function was the minimum utility variance of all nodes along all criteria. The obtained utility values of all nodes were used to calculate the overall utility intervals and ranking of alternatives under evaluation. Numerical example showed that the alternatives could effectively ranked using the proposed interval UTA method, which was compatible with the preference information of decision maker.

The finite-time stability of stochastic Markov jumping systems

ZHOU Shaowei

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  78-84.  doi:10.6040/j.issn.1672-3961.0.2015.362

Abstract ( 1563 )   HTML   PDF (1081KB) ( 669 )   Save

References | Related Articles | Metrics

The finite-time stability of stochastic linear It(^overo)Markov jumping systems was studied. Firstly, the definitions of finite-time stochastic stability and finite-time boundedness were given. Secondly, the necessary and sufficient condition of finite-time stochastic stability was obtained and the state feedback controller design was realized based on linear matrix inequality method. Finally, the validity was verified by a numerical example and system simulation.

Properties of cement stabilized weathered granite material under the influence of soluble salt

YAO Zhanyong, ZHANG Hao, SHANG Qingsen, GONG Benhui, LIU Zhihang, WANG Xugang

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  85-93.  doi:10.6040/j.issn.1672-3961.0.2015.234

Abstract ( 1473 )   HTML   PDF (1971KB) ( 967 )   Save

References | Related Articles | Metrics

In order to analyze the disease mechanism of subbase with cement stabilized weathered granite material and to optimize the material design, the influence of coarse aggregate content on soluble salt corrosion resistance was analyzed through properties tests of three kinds of coarse aggregate content cement stabilized weathered granite material under different curing conditions. Research results indicated that long-term water environment, especially soluble salt existing in water environment, would lead to rapid decay of cement stabilized weathered granite material properties. When coarse aggregate content was less than 20%, the material physical mechanic properties were very poor under the effect of soluble salt, but with coarse aggregate content increasing, material properties improved obviously under various conditions except that cleavage strength reduced slightly under standard curing condition. While coarse aggregate content increased from 20% to 30%, the material worked better in improving compressive ability, water and temperature stability under various environment, as well as tensile property and stiffness under long-term water environment. While coarse aggregate content increased from 30% to 40%, tensile property and stiffness improved more obviously under soluble salt corrosion environment.

Assessment of inter-cavern containment property for underground oil storage caverns using discrete fracture networks

WANG Zhechao, ZHANG Zhenjie, LI Shucai, BI Liping, FANG Shuixin, ZHONG Kecheng

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  94-100.  doi:10.6040/j.issn.1672-3961.0.2015.278

Abstract ( 1542 )   HTML   PDF (3361KB) ( 1040 )   Save

References | Related Articles | Metrics

In order to study the inter-cavern containment property for underground oil storage caverns, combining with Huangdao underground oil storage caverns project, seven different cases were investigated by the discrete fracture network method. The water curtain pressure, inter-cavern oil level difference, cavern spacing and joint inclination angles could influent groundwater head distribution and seepage in fracture network. It was concluded that a higher water curtain pressure, smaller inter-cavern oil level difference or wider cavern spacing could improve the inter-cavern containment property of the facility. The inclination angles of rock joints affected the prevailing flow path in the fracture network and thus the inter-cavern containment property of the facility. It was indicated that a favorable inter-cavern containment property would not be guaranteed by increasing the water curtain pressure alone. And the inter-cavern containment property of an underground oil storage project depended on a combined effect of water curtain pressure, stored oil levels, cavern spacing and joint inclination angle.

The centrifugal model test of cable stress monitoring for rock slope

GUO Yongjian, CAO Zhouyang, SHENG Lijuan

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  101-107.  doi:10.6040/j.issn.1672-3961.0.2015.048

Abstract ( 1545 )   HTML   PDF (2379KB) ( 1210 )   Save

References | Related Articles | Metrics

The centrifugal model test was taken to study the stress monitoring for the three typical types rock slopes,with the similar material mass ratio of quartz sand, plaster, and water as 1.50∶1.00∶0.25∶0.50. Test results showed that stress curves of the three typical types rock slopes were more flatted before the slope failure, stress monitoring and early warning were easy to grasp, and with similar properties,stress curves were used in the form of logistic curves fitting. Stress monitoring site of stratified rock slope should be mainly located in lower part of sliding surface. Failure mode of rock slope controlled by structural plane was mainly in the form of tensile shear,and test stress monitoring site of rock slope should be mainly located in upper part of structure plane,stress monitoring site of cataclastic structure rock slopes should be mainly located in upper part of slope. Centrifuge model test results showed that it was feasible to apply stress monitoring method to evaluate the rock slope stability.

Effect of outside vehicle windshield on aerodynamic performance of high-speed train under crosswind

NIU Jiqiang, LIANG Xifeng, XIONG Xiaohui, LIU Feng

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  108-115.  doi:10.6040/j.issn.1672-3961.0.2015.162

Abstract ( 1662 )   HTML   PDF (5277KB) ( 799 )   Save

References | Related Articles | Metrics

Three-dimensional steady uncompressible Reynolds-averaged N-S(Navier-Stokes)equation and RNG(renormalization group)κ-ε equation turbulence model were used to simulate the aerodynamic performance of the three-carriage high-speed train. The effect of outside vehicle windshield on the aerodynamic performance of high-speed train was studied by change the angle of sideslip. The results showed that testing results generated by the selected algorithm were in agreement with experimental data, and the deviation between them was below 10%. The distribution of the surface pressure coefficient along train was not affected by the windshield, and the amplitude of them had no significant differences. There was significant difference on distribution of the pressure coefficient around the windshield among different cases, and doubled pressure coefficient appeared in some occasions. With the sideslip angle increased, the difference among the six cases was more and more obviously, the maximum difference was 176%. With the increase of sideslip angle, the influence of windshield on lateral force coefficient, lift coefficient and moment coefficient was more significant, maximum differences were 17.71%, 6.35% and 7.52%, respectively. Anti-overturning ability of train with the total enclosed windshield was relatively optimal, and drag force of train with semi typed windshield and smooth typed windshield in wind environment were reduced obviously.

Investigation on temperature characteristic and relaxation mechanism for high-speed train axle-bearing system

LIU Yihui, LI Lin, LIU Kai, SUN Jie

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  116-121.  doi:10.6040/j.issn.1672-3961.0.2015.326

Abstract ( 1605 )   HTML   PDF (2025KB) ( 1017 )   Save

References | Related Articles | Metrics

For the mechanism of axle-bearing relaxation which was easy to occur in high-speed train was revealed and the preventive measures were given. In this paper, the finite element model of temperature characteristics of shaft-bearing under working conditions was established and evaluated with ABAQUS. And bearing components' temperature rise under different speeds was analysized. Combined with the material expansion coefficient test experiment and the experiment for getting matching relations between temperature rise and the bearing inner ring quantity expansion, the temperature field variation characteristics for working components of locomotive were studied. The results of simulation and experiment showed that the amount of interference between the axle and bearing inner ring was reduced with the influence of temperature rise.Finally, the influence between temperature rise and axle-bearing was obtained. The mechanism of relaxation was revealed and the guiding parameters of interference and temperature were given.

Thermoelectric generator system based on engine exhaust gas

MA Zongzheng, SHAO Fengxiang, WANG Xinli, YANG Anjie

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  122-127.  doi:10.6040/j.issn.1672-3961.0.2015.373

Abstract ( 1553 )   HTML   PDF (1784KB) ( 1070 )   Save

References | Related Articles | Metrics

In order to recycle the waste energy from the exhaust gas of the internal combustion engine, one thermoelectric generator system(TEG)based on the TEG technology was experimentally studied. The test bed was added the temperature, voltage and currency measurement system based on the existing engine test systems firstly. Then the impacts of the internal structure of the heat collector on the TEG performance was got by measuring the temperature and noise variation for different inner structure while the impacts of the radiator cooling methods on the TEG performance was reached by measuring the temperature difference of the both ends and output voltage and currency of the TEG. The results showed that the internal structure of the collector with hollow structure was more conducive to maintain the surface temperature uniformity where the temperature difference was under 6 ℃ for the different points which was helpful to electricity energy output, but this structure was not helpful to reduce noise. It also indicated that forced air cooling method was good for improving the TEG temperature difference and enhancing the electricity energy output which could raise the temperature difference by 14 ℃ compared with naturral air cooling method, but the improvement was not changed when the velocity was higher than a limit value, because the heat transfer efficiency was increased simultaneously for both ends of TEG and more energy was released to the environment. Meanwhile, compared with the side cooling method the facade cooling was better which could raise the temperature difference by 15 ℃ except the single face cooling.

Convective heat transfer in micro pin fin heat sinks with different cross section shapes under variable heating loads

GUAN Ning, LUAN Tao, LIU Zhigang, ZHANG Chengwu, JIANG Guilin, QIU Delai

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2016, 46(2):  128-134.  doi:10.6040/j.issn.1672-3961.0.2015.210

Abstract ( 1496 )   HTML   PDF (2494KB) ( 988 )   Save

References | Related Articles | Metrics

In order to investigate the flow and convective heat transfer characteristics under different heating loads in micro pin fins,test sections of micro pin fins with different cross section shapes of circle, diamond and triangle were precision machined. An integraed heating system was used to to measure the pressure drops, frictional resistance factors and thermal resistance in micro pin fins. Based on the experimental results, the impact of heating load on the flow resistance and the heat transfer in micro pin fins with Reynolds number ranging from 0~1 000 was explored. It was found that the friction factors in the three types of micro pin fins became large with the increase of the heating load, but this phenomenon disappeared when Re>400 for the circle and diamond micro pin fins, and Re>250 for the triangular micro pin fins. The convective heat transfer in micro pin fins with cross section shapes of circle, diamond was enhanced by increasing the heating load, but the Nu in triangular micro pin fins became large with the increase of the heating load when Re<250, and then it became slightly smaller when Re>250.