Table of Content

  

20 October 2012
Volume 42 Issue 5

Articles


Articles

Characteristic consistent blocks based rough set in incomplete system

YANG Xi-bei1,2, HUANG Jia-ling1, ZHOU Jun-yi3, YANG Jing-yu2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  1-6. 

Abstract ( 499 )   PDF (1017KB) ( 1735 )   Save

Related Articles | Metrics

The incomplete information system with both absent and missing unknown values was studied. The limitation of characteristic class, was pointed out such that not any two elements in a characteristic class were mutually tolerant. The technique of maximal consistent blocks was introduced into characteristic class, and then the concept of characteristic consistent block was proposed. Not only the basic properties about characteristic consistent blocks based rough set were discussed, but also the relationship between characteristic consistent blocks and characteristic relation based rough sets was analyzed. The results showed that by comparing with the characteristic relation, the greater lower approximation and smaller upper approximation could be obtained by characteristic consistent block approach.

An optimization model for forecasting based on grey system and support vector machine

SHI Jun, ZHU Min

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  7-11. 

Abstract ( 536 )   PDF (1092KB) ( 1311 )   Save

Related Articles | Metrics

Prediction models in traditional gray system involved various factors and fell short in predicting efficiency and precision. An optimized prediction model was put forward by combining the rough theory and the SVM method. The attribute deduction method was first employed on the inconsistent decision table to seek for the core attribute set, which could enable the prediction model to focus better on narrow and specific attribute fields with higher efficiency. A gray model was applied in the optimized dataset. The result parameters were then treated as the input data of a support vector machine for model prediction. China’s census data (1990～2010) were also applied in population prediction. Experimental results showed that this model had better accuracy and higher efficiency than the existing models.

An improved Kalman filter algorithm based on the “current” model

LAN Yi-hua, REN Hao-zheng*, ZHANG Yong, ZHAO Xue-feng

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  12-17. 

Abstract ( 809 )   PDF (1853KB) ( 1713 )   Save

Related Articles | Metrics

An improved Kalman algorithm based on the “current” model was presented to avoid the influence of the acceleration limits. The difference between the velocity forecast estimate and the corrected velocity estimate was utilized to perform adaptive acceleration variance adjustment. The simulation of Kalman algorithms with different acceleration limit parameters proved that the performance of Kalman filter was influenced by the acceleration limits. In addition, the improved Kalman algorithm was compared with standard Kalman filter. The results showed that the proposed method forecast more accurately than the standard Kalman filter.

The automatic extraction of the chemical bonds information in the chemical structure images

SUN Lan-lan1,2, LI Cun-hua2*, GUAN Yan2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  18-23. 

Abstract ( 818 )   PDF (1494KB) ( 1825 )   Save

Related Articles | Metrics

The inflection points are very important for chemical structure graphs. If the inflection points of chemical structures could be discriminant effectively, they would make the effectiveness of the extraction of chemical bond information improved greatly. The characteristics of the chemical structure was analyzed and summarized, and the advantages and disadvantages of the Hough method was balanced. It was found that the offset value of the edge points was with strong regularity,which played an important role on the extraction of the inflection points and the judgment of chemical bond type. Finally, the experiments selected 100 BMP figures of the molecular structure to extract and count the chemical bond information. The statistic results showed that the accuracy rate of chemical bonds information extraction was of 7383％， which proved the effectiveness of this method.

A TOA estimation algorithm based on envelope extraction

LIU Qian1, XIA Bin1*, PENG Rong-qun1, CHEN Nai-shu2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  24-29. 

Abstract ( 690 )   PDF (2905KB) ( 1321 )   Save

Related Articles | Metrics

The noise could affect the location precision in ultra wideband wireless sensor network. In order to solve this problem, a time of arrival (TOA) estimation algorithm based on envelope extraction was proposed. The noise component of signal could be effectively removed by the multiresolution analysis of wavelet transform. Then the envelope of denoising signal was extracted by Hilbert transform. Finally, the maximum value of the first envelope was used as the TOA estimation value. The simulation results showed that the algorithm could suppress the noise and improve estimation precision.

Blind data processing in cloud computing based on trusted computing mechanisms

JI Tao, LI Yong-zhong

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  30-34. 

Abstract ( 641 )   PDF (1621KB) ( 1440 )   Save

Related Articles | Metrics

Aimed at solving the problem that sensitive data was subjected to unauthorized access and illegal tampering during data processing in cloud computing environment, a new approach called blind data processing based on trusted computing mechanisms was proposed. First, the root of trust in cloud computing environment was established by using trusted platform module(TPM) to ensure that the sensitive data was bound to a particular state of a cloud computing system. Then, the blind data processing environment was constructed. The messages transmitted over the cloud were encrypted by using TPM. The system integrity was measured and the remote attestation was carried out. Data migration from the source to the target side was completed by using elliptic curve cryptographic algorithms. The analysis showed that the secure execution environment for data processing in cloud computing was provided by the proposed approach.

Research on Web negative information mining based on event ontology

LIU Dong-hui1,2, JIANG Wei1*

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  35-40. 

Abstract ( 470 )   PDF (1865KB) ( 1334 )   Save

Related Articles | Metrics

In order to mine the negative information on the internet, the eventbased semantic analysis technology was applied. The methods of event ontologybased Web negative information mining, especially event ontology construction and text feature reconstitution were studied. Information clustering was used as an example to validate the methods proposed. A prototype system based on event ontology was implemented. Experimental results showed that the average accuracy obtained by the event ontologybased and kmeans method was 721％, which increased by 53％ compared with the traditional kmeans clustering method.

Research on intrusion detection algorithm based on PCA and semisupervised clustering

DING Yan, LI Yong-zhong*

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  41-46. 

Abstract ( 616 )   PDF (1093KB) ( 1480 )   Save

Related Articles | Metrics

In order to solve the problem that lots of redundant information existed in network intrusion detection data and the traditional clustering algorithms were inadequate for detecting outlier, an intrusion detection algorithm based on principal component analysis(PCA) and semisupervised clustering was proposed. First, the features of data were extracted by using PCA, and the redundant attributes among the data were eliminated. Then, a few labeled samples and pairwise constraints information were exploited, and competitive agglomeration was introduced to letting the system active learning in order that the detection of lots of unknown samples could be realized. The experimental results on intrusion detection data set and UCI benchmark data sets showed that this algorithm could effectively improve the system performance.

Flame detection based on LBP features with multiscales and SVM

YAN Yun-yang1,2, TANG Yan-yan2, LIU Yi-an2, ZHANG Tian-yi3

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  47-52. 

Abstract ( 810 )   PDF (1665KB) ( 3078 )   Save

Related Articles | Metrics

Fire detection based on videos is an effective method to prevent fire in large spaces. The texture of flame is special. Multiscale texture features were extracted to improve the flame detection performance due to its much more discrimination information. The flame candidates were located by character of flame brightness at first. Then different patterns of LBPfeature with different scales were extracted from these candidate areas. Finally, these features were put into SVM classification to recognize whether it was a flame or not. Experimental results showed that the method had a simple computation and could accurately recognize flame in video sequences and the false positive was low.

Price forecasting model based on linear backfilling and adaptive sliding windows

ZHU Quan-yin1, YAN Yun-yang1, ZHOU Pei1, GU Tian-feng2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  53-58. 

Abstract ( 728 )   PDF (1711KB) ( 1462 )   Save

Related Articles | Metrics

The accuracy rate of commodities price forecast based on Web mining is lower because of the network noise. In order to increase this accuracy rate, a novel price forecast method and a comprehensive price forecast model based on the linear backfilling and adaptive sliding windows algorithm were proposed. This comprehensive price forecast model was utilized in the commodities price forecast for cell phone and gold market. Experimental results showed that the mean absolute error of this proposed model could get more than 99 percent accuracy rate. In addition, the antinoise performance of the webpage commodity price data extraction was improved. At the same time,this method could also solve the problem that the existing vendors only had the historical sales price data but did not have the forecasted price based on a plurality of vendors, which could also provide basis for the commodities market forecast and analysis.

Image normalization based on local autocorrelation and its application to face detection

ZHU Hong-jin1, FAN Hong-hui1, CHEN Xing-rui1, TAMURA-Yasutaka2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  59-64. 

Abstract ( 579 )   PDF (1668KB) ( 1165 )   Save

Related Articles | Metrics

Nonuniformity of luminance in images due to irregular lighting etc. could cause difficulties in various kinds of image processing in face detection. A normalization method was presented for recognizing human faces under variation in lighting, which was called local autocorrelation (LAC) method. LAC method was applied to human face detection based on Adaboost algorithm. The classification result of CMU PIE database for original and LAC images were compared with the LAC method. The physical properties of the LAC were analyzed, and the LAC robustness of linear changes in illumination was verified theoretically. Experimental results showed the number of weak classifiers could be reduced to a great extent, while preserving equal detection capability. The effectiveness of elimination of nonuniform illumination variation in images was verified in face detection experiment.

Segmentation algorithm of chemical molecular structure images

GUAN Yan, LI Cun-hua*, ZHONG Zhao-man, SUN Lan-lan

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  65-70. 

Abstract ( 943 )   PDF (1446KB) ( 1504 )   Save

Related Articles | Metrics

The segmentation algorithm of chemical molecular structure based on area size and bending degree was proposed to segment chemical bonds, heteroatoms and perssad. First, chemical molecular structure images were segmented into two parts according to connection area size. One was the combination of all big size connection areas containing chemical bonds, and the other was the combination of heteroatoms, perssad and single chemical bonds. Second, single lines were extracted based on bending degree, which represents small size combination images, and likelines such as “I”,“l” and “-”. Finally, “I”,“l”, “-” and single lines were distinguished, chemical bonds of single lines and big size connection areas were combined, and the segmentation of chemical bonds, heteroatoms and perssad was realized. The accuracy by the proposed algorithm reached to 983％, and the segmentation effect was consistent with human visual perception. This is the foundation for automatically extracting chemical molecular structure images.

A novel method for face recognition based on generalized rotation invariant kernel

GUO Hui-ling, WANG Shi-tong*, YAN Xiao-bo

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  71-79. 

Abstract ( 605 )   PDF (1630KB) ( 1602 )   Save

Related Articles | Metrics

Rotation invariant kernel was applied to face recognition and some other areas, but its antinoise ability was unsatisfied. Elide the specific form of distribution, the generalized rotation invariant kernel, which can convert the original nonlinear problem into linear one, was introduced. Meanwhile, parameter estimated difficulty was reduced. The index α of generalized Gaussian function played a decisive role on the peak. With reference to this property, index r was introduced into the algorithm. Through controlling the change of index r, recognition rate of the algorithm was observed. And the antinoise ability of the algorithm could be proved by adding different Gaussian white noise into the experimental data.Experimental results proved the superiority. Recognition rate was almost linearly changed while the index r changed and the best r always existed for the best recognition rate, which was better than that of rotation invariant kernels.Under the same experimental conditions,the antinoise ability was greatly improved.

Water quality prediction model based on APSO-WLSSVR

XU Long-qin1, LIU Shuang-yin1,2,3,4*

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  80-86. 

Abstract ( 667 )   PDF (1761KB) ( 1588 )   Save

Related Articles | Metrics

In order to solve the problem of low prediction accuracy, the bad robustness of the traditional forecasting methods and the standard least squares support vector regression (LSSVR) in water quality prediction, the adaptive particle swarm optimization weighted least squares support vector regression (APSO-WLSSVR) model for water quality prediction was proposed. Different weights were set for various samples according to its different importance for the model. A weighted least squares support vector regression model (WLSSVR) was established, which could avoid low prediction accuracy of the standard LSSVR, and ignore the importance of the samples differences. The particle swarm optimization algorithm was adopted to optimize and choose the model parameters to reduce the blindness and the impact of human factors of the standard LSSVR trial method when obtaining the parameters. In order to verify the performance of the model, the water quality of intensive farming river crab in Yixing, Jiangsu Province, was predicted, which was also compared with other forecast methods. The results showed that the model prediction accuracy was obviously improved, and also had good robustness and generalization ability, which could met the practical needs of the intensive aquaculture water quality management.

A spam short message classification method based on word contribution

ZHANG Yong-jun1, LIU Jin-ling2, YU Chang-hui3

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  87-90. 

Abstract ( 753 )   PDF (1003KB) ( 1575 )   Save

Related Articles | Metrics

A classification method based on word contribution was proposed to classify spam short messages. The concept of word contribution was introduced for representing weight difference of a word in different categories, the word contributionclassification matrix was constructed, then the mean square deviation of each row in the matrix was computed to reduce dimensionalities. To determine the classification a short message belongs to, short messagecategory membership degrees were calculated based on word contribution. Furthermore if category candidates were more than one, the classification conflict problem could be resolved by comparing the densities of short messagecategory membership degree. The experimental results showed that the proposed method was superior to other classification methods in the classification result and realtime.

The technique of gas disaster information feature extraction based on rough set theory

LI Hui1,2, HU Yun1,3, LI Cun-hua1

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  91-95. 

Abstract ( 589 )   PDF (1101KB) ( 1430 )   Save

Related Articles | Metrics

In order to accurately predict coal and gas outburst danger and to establish an effective earlywarming support system of gas in coal mine, a high efficient gas disaster feature extraction algorithm based on rough set was proposed in view of the characteristics of coal mine gas disaster. The algorithm first refined the gas disaster information matrix by using dimensionality reduction, then the entropy and max entropy in the concept of rough set theory were used to establish data mining model of gas disaster prediction. The effectiveness and practicality of rough set theory in the prediction of gas disaster and feature extraction was confirmed through practical application.

An algorithm for protecting location privacy in road network

SUN Lan, LUO Zhao, WU Ying-jie, WANG Yi-lei

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  96-101. 

Abstract ( 663 )   PDF (1616KB) ( 1420 )   Save

Related Articles | Metrics

Recently, the privacy preserving locationbased services has been a hot topic in data privacy preserving research fields. The existed researches on location privacy preserving mainly focused on Euclidean space. However, many locationbased services were under roadnetwork environment, whose distribution of users was possibly unbalanced, which could make the traditional location privacy models and methods under Euclidean space unusable. A location privacy protection algorithm was proposed to prevent the inferring attack caused by the unbalanced distribution of users in roadnetwork. The key idea of the proposed algorithm was that the cloaked segment set was constructed by sorting edges with edge weight and taking the geographical position distribution of users into consideration. Experimental analysis was designed by comparing the proposed algorithm and the traditional algorithm on the feasibility and effectiveness. Experimental results showed that the proposed algorithm was effective and feasible.

Research on advanced prediction and surrounding rock stability of shallow buried and unsymmetrical loaded section in tunnel excavation

Lü Guo-ren1, ZHANG Shou-long2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  102-107. 

Abstract ( 620 )   PDF (2535KB) ( 1292 )   Save

Related Articles | Metrics

In order to study the surrounding rock stability of shallow buried and unsymmetrical loaded section, the geological radar was used to perform advanced geology forecast by taking the construction of Xiejiahe Tunnel of Yanhai Highway as an example. Based on the forecast results, the construction program was optimized and the leading ductule injecting was adopted. The site monitoring was also employed and the results demonstrated that the data were stable and the construction program was feasible. The shallow buried and unsymmetrical section of the Xiejiahe tunnel was stable enough to assure tunnel construction going on smoothly. Furthermore, it could provide reference for similar engineerings.

Analysis on influence factors of the earlyage cracking of concrete pavement

WANG Hao1, LI Peng-cheng2*, MAO Hong-lu2, SUN Ren-juan2, GE Zhi2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  108-112. 

Abstract ( 729 )   PDF (1799KB) ( 1342 )   Save

Related Articles | Metrics

The software HIPERPAV was adopted to analyze the major influential factors of the earlyage cracking of concrete pavement, which including the slab thickness, fly ash, curing method and time, construction time, paving season and time of sawing. The results showed that the stressstrength ratio of slab constructed in summer was lower than that in winter, and the pavement slab without saw cutting was more likely to produce the earlyage random cracking than the slab with cutting. In addition, the proper mixture ratio, construction time and curing method could significantly decrease or prevent the earlystage cracking.

Preparation and characterization of honeycomb microporous films based on reversible addition fragmentation chain transfer (RAFT) polymerization

WANG Li-ping1,2, Lü Xin-hu2*, LI Guang 2, DENG Ai-xia2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  113-117. 

Abstract ( 638 )   PDF (2087KB) ( 1766 )   Save

Related Articles | Metrics

The polystyrene (PS) was synthesized via reversible addition fragmentation chain transfer (RAFT) radical polymerization using azobisisobutryonitrile (AIBN) as the initiator, (S, S)′-bis (α, α)′-dimethyl-α″-acetic acid) (trithiocarbionate) (BDAAT) as the chain transfer agent (CTA). In addition, the polystyrene (PS) was synthesized by traditional radical bulk polymerization. Under the high humidity and nitrogen conditions, the microporous films were prepared by water drop templated method using polystyrene (PS) and polyethylene glycol (PEG) as materials of the membranes respectively. The mechanism of the formation of PS honeycomb microporous films was discussed. Moreover, the influence of the polymerization manner, the solution concentration, and the volatility of solvents on the structure of the microporous films were also studied. The results showed that the morphology of microporous films was affected by the end group of linear polymer, hydrophilic matter and so on. The diameter of film pore decreased obviously with the increasing of the polymer concentration.

Stucture optimization of high power LED heat sinks based on the method of minimum entropy generation

LAI Yan-hua, WEI Lu-lu, Lü Ming-xin, ZHAO Lin-yan, YUE Hong, LIU Cun-fang

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  118-122. 

Abstract ( 675 )   PDF (1639KB) ( 2069 )   Save

Related Articles | Metrics

In order to enhance the heat transfer of heat sink with straight fins of the high power LED lamps working under natural convection, the structure optimization of this kind of heat sinks was carried out. Based on a method of the minimum entropy generation, the influences of the important parameters on the heat transfer and fluid friction of the heat sink were analyzed, such as the fin thickness and the numbers, the height of fin, total quality of the heat sink and wind speed. The results showed that the dimensionless entropy generation and the temperature entropy generation first increased, and then decreased with the number of fins increasing, and the entropy generation caused by temperature was about ten times of that by friction. In addition, under a given heat flux, there existed the best geometry parameters of heat sink fins, which could minimize the entropy generation, and could improve the whole performance of the heat sinks.

Research on energy absorption and thermal characteristics of MOV in fault current limiter

TANG Zong-hua1,2, TAN Zhen-yu2, WEN Hui2, SUN Shu-min1

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  123-129. 

Abstract ( 746 )   PDF (2060KB) ( 1895 )   Save

Related Articles | Metrics

As a key part of series resonant type fault current limiter, the reliable operation of metal oxide voltage limiter(MOV)is especially important. By using electromagnetic transient analysis program, a 110 kV single-phase circuit was taken as the example for analyzing the energy absorption characteristics of MOV for the system short-circuit fault. The results showed that the longer short-circuit faults time, the more energy that the MOV absorbed and that the closed time of fast switch could determines the energy that MOV absorbed. In addition, the ANSYS finite element analysis program was also used for analyzing the transient thermal characteristics of MOV, by taking MOV valve units and the whole MOV as the research object. The temperature rise of MOV valve units as a function of the time of short-circuit fault and the injected energy was obtained. At the same time, the temperature field distribution of MOV valve units and the whole MOV were worked out, the temperature distribution characteristics were also analyzed. Based on the above analyses, a practical MOV was constructed and its temperature increase was experimentally determined. Good agreement between the determined temperature increase and the corresponding theoretical calculations was obtained, which indicating the calculations reliability. The research results could provided theoretical direction for the design of configuration strategy for fault current limiters and for the protection strategy of MOV temperature rise, and could present theoretical reference for the design of MOV.

Cloud droplets obtaining algorithm based on normal interval number

LI Xiu-hai1, YU Shao-wei2*

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2012, 42(5):  130-134. 

Abstract ( 651 )   PDF (1394KB) ( 1537 )   Save

Related Articles | Metrics

Cloud droplets are difficult to be obtained in constructing backward cloud. To slove the problem, a new algorithm using normal interval number was proposed by analyzing the traditional algorithms of backward cloud and in the inspiration of the idea that all fuzzy membership functions could form a curve cluster which could be regarded as an approximation to a cloud.The new algorithm could obtain the cloud droplets by generating some normal numbers in expectation curve and substituting them in each membership function.Experimental results showed that the algorithm could get abundant cloud droplets,which could meet the distribution of normal cloud model.