Loading...

Table of Content

      
    20 February 2013
    Volume 43 Issue 1
    Articles
    An improved shuffled frog leaping algorithm for solving constrained optimization problems
    ZHANG Xiao-dan, ZHAO Li, ZOU Cai-rong*
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  1-8.  doi:10.6040/j.issn.1672-3961.0.2012.002
    Abstract ( 938 )   PDF (1417KB) ( 2121 )   Save
    Related Articles | Metrics

    An improved shuffled frog leaping algorithm (Im-SFLA) was proposed for solving constrained optimization problems. In view of overcoming the defects of shuffled frog leaping algorithm (SFLA) such as slow searching speed in the late evolution and local minimum, the ideas of simulated annealing and immune vaccination were involved into basic SFLA with Guassian mutation and chaotic disturbance in the improved algorithm. The test results on standard test functions indicated that Im-SFLA could outstandingly enhance the convergence velocity and precision, effectively averted the local extreme values and the global searching performance was superior to SFLA. The static penalty function was used to transform a constrained optimization problem into an unconstrained optimization problem, and the test results on 12 constrained optimization benchmark functions showed that Im-SFLA could obtain a high solution quality and had strong robust, which was an effective algorithm for solving constrained optimization problems.

    A modified general regression neural network with its application in traffic prediction
    YI Liang-zhong1, ZHANG Chao2*, PEI Zheng3
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  9-14. 
    Abstract ( 856 )   PDF (1587KB) ( 1318 )   Save
    Related Articles | Metrics

    Based on the method of k nearest neighbors algorithm, the optimum smoothing parameter was found by means of network performance evaluation. The approach depended not only on the value of mean square error, but also could sort the mean square error without affecting the forcasting performance. The optimum smoothing parameter was difficult to be found because of the volatility of the data solved by the modified algorithm. Finally, a traffic forcasting experiment was provided to analyze the effectiveness of the proposed algorithm. The results revealed that the optimum smoothing parameter found by means of k nearest neighbors could obtain the minimum prediction error.

    New scalable ROI algorithm based on visual attention
    CHEN Xu1, ZHANG Ji-hong2, LIU Wei2, LIANG Yong-sheng2
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  15-21. 
    Abstract ( 691 )   PDF (2779KB) ( 2284 )   Save
    Related Articles | Metrics

    In order to make full use of video coding bit-steam and high subjective QoE(quality of experience), a AOF(area of fixation) detection model was proposed based on current achievements of psychology and visual attention. Combined with saliency calculation and eye-movement tracking analysis, this model could obtain the AOF of different viewing duration. According to the AOF, a method of content-aware rate adaptation was also proposed.  The experimental results showed that the proposed method could outperform than the other existing methods and could improve the subjective QoE when coding and transmission in situation of limited bandwidth.

    Web spam detection based on SMOTE and random forests
    FANG Xiao-nan1,2, ZHANG Hua-xiang1,2*, GAO Shuang1,2
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  22-27. 
    Abstract ( 981 )   PDF (1009KB) ( 3711 )   Save
    Related Articles | Metrics

    Web spam refers to the actions intended to mislead search engines into ranking some pages higher than they deserved, which could significantly deteriorate the quality of searching results. Considering the serious imbalance of the Web spam dataset, it was proposed to use oversampling method SMOTE to balance the dataset, then to train the classifiers with random forests algorithm. The results showed that the SMOTE+RF method was more prominent by means of experimental comparison with the conventional single classifiers and the ensemble learning classifiers. The important parameters of this method were optimized based on experimental results, and the reasons for the improvement of the AUC value after using SMOTE were also analyzed. Experimental results on WEBSPAM UK2007 dataset showed that this method could markedly improve the performance of the classifiers, of which the AUC value could exceed the best result of Web Spam Challenge 2008.

    Text categorization algorithm based on non-linear manifold learning and k-NN
    ZHANG Guo-dong1,2, ZHANG Hua-xiang1,2*
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  28-33. 
    Abstract ( 735 )   PDF (882KB) ( 1197 )   Save
    Related Articles | Metrics

    In order to save the problems of dimensionality curse, noise data in text categorization, the text categorization algorithm was presented based on the non-linear dimensionality reduction algorithm and combined with kNN(knearest neighbor algorithm). The algorithm first removed the noise data, and then used the locally linear embedding algorithm of non-linear manifold learning to recover low-dimensional manifold structure in high-dimensional data to implement dimensionality reduction. The processed data was used to construct k-NN classifiers. Experimental results showed that this  algorithm could  effectively improve the accuracy of text classification.

    A clustering method based on 18-element linguistic-valued fuzzy similar matrix
    ZHANG Yun-xia, CUI Xiao-song, ZOU Li*
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  34-40. 
    Abstract ( 757 )   PDF (991KB) ( 1429 )   Save
    Related Articles | Metrics

    Based on the 18-element linguistic-valued lattice implication algebra, the linguistic-valued fuzzy similarity of linguistic-valued fuzzy set was proposed. In addition, the calculating formular for average linguistic-valued fuzzy similarity was given and its reasonability was proved. Furthermore, the linguistic-valued fuzzy similar matrix, linguistic-valued fuzzy equivalent matrix and λ-cut matrix of linguistic-valued fuzzy matrix were constructed. Finally, a new fuzzy clustering method based on linguistic-valued fuzzy similar matrix was proposed, and an example was given to illustrate its effectivity in fuzzy clustering problem.

    A fast algorithm for color space transformation based on SSLUT
    LIU Yi-fang1,2, ZHANG Yun-feng1,2*, CHI Jing1,2, ZHANG Cai-ming1,2
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  41-47. 
    Abstract ( 812 )   PDF (1357KB) ( 2098 )   Save
    Related Articles | Metrics

    To meet the efficiency requirement of color space conversion in real-time video systems, a new fast algorithm for the color space YCbCr to HSV conversion was proposed based on the simplified shift and look-up table (SSLUT). The transformed image quality assessment was implemented. The process of color space conversion was divided into two steps. First, the fixed-point shift method was used to convert YCbCr to RGB. Second, the color space RGB to HSV conversion was used for look-up table method. The fast algorithm speed was tested based on DSP and PC Testing Platform. The image quality assessment was achieved using the structural similarity (SSIM) and the peak signal to noise ratio (PSNR). The results showed that the real-time capability was satisfactory and the transformed image had high fidelity.

    Conflict evidence in trust model based on evidence theory
    ZHANG Xin-yi, ZHAI Yu-qing*
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  48-53. 
    Abstract ( 773 )   PDF (1112KB) ( 1235 )   Save
    Related Articles | Metrics

    The trust model based on evidence theory was a widely used trust model, and the evidence theory was an uncertainty reasoning method with solid theoretical foundation. In order to slove the problem of the evidence collection and the unreasonable results made by conflict evidence combination, the method of the evidence collection was modified to enhance the quality of evidences,  and the conflict evidences were redistributed based on distance, similarity, support and credibility of evidences during the process of evidence combination. Experimental results showed that the improved combination mechanism of evidences could enhance the reasonality of the evidence combination and could make the results of trust computing more credible.

    Application of random variable fuzzy recognition model in  cleaner production assessment
    WANG Lin1, LIU Bao-dong2 *
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  54-62. 
    Abstract ( 654 )   PDF (976KB) ( 1477 )   Save
    Related Articles | Metrics

    The random variable fuzzy recognition model was established for the coexistence of randomness and fuzziness in cleaner production assessment. A random term obeying normal distribution was introduced to the variable fuzzy recognition model, so the randomness and fuzziness could be studied simultaneously. Four random fuzzy recognition sub-models were generated by controlling the distance parameter p and the optimization parameter α, and then according to the 3σ principle of normal distribution, after choosing a reasonable confidence, the grade was identified by the confidence interval of grade feature value. Compared with the scoring model of centesimal system, this model could cut down the effect of subjective preference on assessment results, and could improve the maneuverability and accuracy in evaluation process. The experimental results showed that the random variable fuzzy recognition model could evaluate the cleaner production level accurately, which could also provide a reference for cleaner production assessment.

    A multigranulation rough interval-valued fuzzy set approach
    FAN Wei
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  63-68. 
    Abstract ( 726 )   PDF (866KB) ( 1064 )   Save
    Related Articles | Metrics

    Multigranulation was an effective approach for dealing with distributed data by rough set. To approximate the fuzzy target by multigranulation approach, a multigranulation rough interval-valued fuzzy approach was proposed, which included optimistic and pessimistic models. Not only the basic properties of such two different multigranulation rough interval-valued fuzzy sets were discussed, but also an illustrative example was analyzed to show the effectiveness of the proposed models.

    Robust L2-L∞ control of  δ  operator formulated time-delay systems with disc regional pole constraint
    HE Rong-fu1, XIAO Min-qing2*
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  69-79. 
    Abstract ( 544 )   PDF (1544KB) ( 1546 )   Save
    Related Articles | Metrics

    The problem of robust L2-L∞ control with disc regional pole constraint was investigated for a class of δ operator formulated linear time-delay uncertain systems. Based on the  matrix theory, a sufficient condition for the poles of a  δ  operator formulated linear time-delay system, being inside a disc region, was presented in the form of linear matrix inequality (LMI). For a class of  δ  operator time-delay uncertain systems with disturbances, a design method of the optimal robust L2-L∞ controller was obtained by the Lyapunov approach, which could place the poles of the closedloop system in a prespecified disc and ensure the L2-L∞ performance to be optimal, . A numerical example was provided to illustrate the feasibility and effectiveness of the design method.

    Research on pressure sensitivity of the conductive asphalt mortar  with carbon fiber and graphite powders
    YAO Zhan-yong, HAN Jie*, SHANG Qing-sen, GE Zhi, ZHANG Xiao-meng, CUI Heng
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  80-85. 
    Abstract ( 583 )   PDF (1666KB) ( 1986 )   Save
    Related Articles | Metrics

    Conductive components were prepared by carbon fiber and graphite powders, and the relationships between different dosage of carbon fiber, graphite powders and asphalt mortar resistivity were investigated by aboratory test. The material pressure sensitivity and pressure-sensitive reversibility were analyzed by cyclic compressive loading test. The change rate of electrical volume resistivity Fn and the fractional change in resistivity per unit stress Kn were analyzed. The regularity and trend of variations of Fn and Kn under different loading circles were discussed by the mathematical statistics method. The results showed that the carbon fiber and graphite powders could reduce the resistivity of the asphalt mortar effectively. The conductivity increased with the compression stress and decreased with the compression stress increasing unloading, which showed good pressure sensitivity. There was a good correlation between Fn, Kn and numbers of cyclic compressive loading. With the loading cycles increasing, the change in resistivity per unit stress decreased, demonstrating that the reversibility of material tend to be stable. Reducing plastic deformation is very important to the reliability of initial detection results.

    Study on water-inrush prediction of coal floor based on the limit equilibrium theory of rock mass
    ZHANG Wei-jie1, LI Shu-cai1, WEI Jiu-chuan2, ZHANG Qing-song1
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  86-91. 
    Abstract ( 683 )   PDF (1981KB) ( 1653 )   Save
    Related Articles | Metrics

    Based on the limit equilibrium theory of rock mass, the mechanical mechanism of water-inrush of seam floor was analyzed. The theoretical formulae of the limit hydraulic pressure of seam floor were deduced through taking the facing length,the thickness and the lithological association of floor rockmass into consideration. The water-inrush risk was predicted according to the geology condition in Baode coal mine. By contrasting and analyzing, the method of limit hydraulic pressure of seam floor could improve the water inrush coefficient method, and could be more effective in predicting the water inrush danger with bigger facing length.

    Feasibility of using magnetic resonance sounding in prediction of water bearing structures in front of a tunnel face
    SUN Huai-feng1, LI Shu-cai1, LI Xiu2, QI Zhi-peng2, LIU Lei2, XUE Yi-guo1, SU Mao-xin1, LIU Bin1, ZHANG Wen-jun1
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  92-97. 
    Abstract ( 777 )   PDF (1425KB) ( 1590 )   Save
    Related Articles | Metrics

    The use of magnetic resonance sounding in prediction of water bearing structures in front of a tunnel face was proposed in order to realize the effective detection of water in front of a tunnel face. A homogeneous conductive full space was used as the basic model. The distribution of harmonic electromagnetic response was calculated. Distribution of the magnetic resonance toppled angles in magnetic resonance sounding(MRS) detection on tunnel face was also simulated. The modeling result showed the obvious MRS response from the water in front of a tunnel face, while the amplitude was very small. According to the characteristics of MRS detection in tunnels, it was proposed that the currently cosine excitation source could be replaced with step source and the current receiver coil could be replaced with magnetic probe. MRS is a potential method to detect water and distinguish water and mud.

    The oretical analysis on application of regional controlled recirculation ventilation of Nanwenhe River Tungsten Mining
    LI Jin-feng1, XIE Xian-ping1, BAO Wen-jun2, WU Yang3
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  98-103. 
    Abstract ( 694 )   PDF (947KB) ( 1604 )   Save
    Related Articles | Metrics

    By using the controlled circulating ventilation to optimize and transform, the problems of weak ventilation, low efficiency of effective air volume, disordered layout of local fan in 1232m midpiece in south side in Mining plant of Nanwenhe river tungsten mining was analyzed. Based on reasonablely selecting parameters of recirculation rate, recirculation air volume and purifying efficiency, it was found that, in the situation, the controlled circulating ventilation could make the air volume increase to 84.66m3/s from 49.45m3/s at the working face that effective dowmcast air keep stable. In traditional ventilation system, the powder consumed increased 5.021 times when air quantity at site increased  to 84.665m3/s from 49.45m3/s, comparatively, when controlled circulating ventilation was used, powder consumed was increased just 1.845 times. In terms of a mining area of capacity of 450 000 tons per year, it was of significance to apply controlled recirculation ventilation technology, which could be expected to obtain good economic benefit.

    Preparation of SiO2-ZrO2 compound high temperature adhesive
    HE Zhao-pin, SHI Yuan-chang*, SUN Li-bo, LI Bo, YUAN Ye, LIU Jiu-rong
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  104-108. 
    Abstract ( 809 )   PDF (1766KB) ( 3203 )   Save
    Related Articles | Metrics

    In order to improve the high temperature resistant performance of traditional silicate inorganic adhesive, the new-type compound high temperature adhesive was prepared by sol-gel method with tetraethyl orthosilicate and zirconium oxychloride. Its thermal stability and bonding performance were studied by characterizing the structure, phase composition, heat change and appearance. The results showed that pH was of great influence on the sol-gel reaction process, and the optimal pH was 3. The preparation methods had obvious influence on the microstructure of SiO2-ZrO2 compound adhesive. Dispersion method was better than the in situ synthesis method for preparing the compound adhesive. The molar ratio of raw materials was an important influencing factor on the performance of the compound adhesive. The SiO2-ZrO2 compound adhesive could provide good bonding performance and good thermal stability when the Si-Zr molar ratio was 1∶3.

    Simulation for separation of ethyl acetate and ethanol by heat-integrated pressure swing distillation
    YANG Jin-bei1, 3, YU Mei-qiong1, ZHENG Zhi-gong2, QIU Ting3*
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  109-114. 
    Abstract ( 708 )   PDF (2044KB) ( 2080 )   Save
    Related Articles | Metrics

    Heat-integrated pressure swing distillation for separation of ethyl acetate and ethanol was investigated based on the pressure-sensitivity of the azeotropic composition, and the separation sequence was formed by two columns operating at different pressures. The simulation of separation process was carried out satisfactorily by means of a package of commercial software Aspen Plus and using the thermodynamic model NRTL (non random two liquids) with binary parameters between ethyl acetate and ethanol obtained by experimental data of vapor liquid equilibrium (VLE). The factors affecting the separation efficiency and energy consumption including the number of theoretical plates, feed location and reflux ratio of two columns were analyzed. The results showed that ethyl acetate and ethanol could be efficiently separated by heat-integrated pressure swing distillation, and the optimal conditions were obtained. The heat-integrated pressure swing distillation could save heating steam by 34.7% compared with conventional pressure swing distillation, which could provide an effective method for the design and energy-saving of azeotropic system separation.

    Linguistic interval 2-tuple power average operator and its application
    YAN Li, RUAN Yan-li, PEI Zheng*
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  115-122. 
    Abstract ( 694 )   PDF (915KB) ( 2014 )   Save
    Related Articles | Metrics

    Due to the uncertainty of decision making, a linguistic interval 2-tuple representation and a linguistic interval 2-tuple power average (LI2TPA) operator were proposed, and some desired properties of the developed operator were also studied. Because the support measure between two aggregated objects was analyzed in LI2TPA operator, this operator could relieve the influence of unduly high or unduly low evaluation values on the collective linguistic evaluation result. A numerical example showed the effectiveness of the new operator in linguistic decision making.

    Thermally driven in porous medium in a centrifugal force field
    XIA Jie1,2, CHANG Hai-ping1*
    JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2013, 43(1):  123-126. 
    Abstract ( 512 )   PDF (1690KB) ( 1112 )   Save
    Related Articles | Metrics

    Based on the research of turbine blade super cooling technology, the porous medium was installed in a new kind of cooling configuration with cooling tunnels. Numerical simulations were carried out to investigate the thermally driven in the new kind of cooling configuration in a centrifugal force field. The results of experiments and numerical simulations were basically consistent. The results showed that the thermally driven happened after porous medium was installed in a new kind of cooling configuration. Furthermore, the fluid and heat transfer processes were found in enclosure through fluid and temperature field gained by numerical simulations.