Table of Content

  

20 October 2015
Volume 45 Issue 5

Discriminative manifold-based uncorrelated sparse projective nonnegative matrix factorization

LI Xinyu, XU Guiyun, REN Shijin, YANG Maoyun

Journal of Shandong University(Engineering Science). 2015, 45(5):  1-12.  doi:10.6040/j.issn.1672-3961.2.2014.155

Abstract ( 1343 )   PDF (4510KB) ( 1044 )   Save

References | Related Articles | Metrics

Inspired by manifold learning, sparse representation and discriminant analysis theories, a discriminative manifold-based uncorrelated sparse projective nonnegative matrix factorization(DMUPNMF)algorithm was developed in this work. By exploiting local and nonlocal geometric discriminant information of the data and the merits of the linear projective NMF, the extracted features were approximately uncorrelated and the decomposition results of DMUPNMF were sparse and better part-based representation. Multiplicative updating rules were introduced to slove the optimization problem of DMUPNMF and its convergence proof was given as well. Moreover, projected gradient decent optimization method was developed to enhance the convergence speed of the method. An approach was proposed to select the informative data points from training dataset, which reduces the computation burden and storage space resulted from a large amount of objects for traditional NMF. Simulations demonstrated that the proposed algorithm outperforms the state-of-the-art algorithms on real-world problems.

Image classification algorithm based on minimax probability machine with regularized probability density concensus

WANG Xiaochu, WANG Shitong, BAO Fang

Journal of Shandong University(Engineering Science). 2015, 45(5):  13-21.  doi:10.6040/j.issn.1672-3961.2.2015.168

Abstract ( 950 )   PDF (6221KB) ( 335 )   Save

References | Related Articles | Metrics

In order to solve image classification problem of which the images contained labeled and unlabeled samples, this research proposed an image classification algorithm based on minimax probability machine regularized by probability density concensus(called PDMPM). The distribution of the image samples in the hyperplane was estimated by using the probability density estimation function and probability density estimation constrained item was got by minimizing the distribution of the labeled and unlabeled samples. The probability density estimation constraint item was integrated into the nonlinear minimax maximum probability machine and used for image classification. The accuracy of the proposed algorithm was increased by 3.99% compared with Gaussian kernel minimax probability machine in the test of Yale face database, five of the Caltech 101 database and ten of Fifteen Scene Categories Dataset. Experimental results indicated that the method made full use of the distribution information of unlabeled image samples and made the image classification hyperplane more accurate.

Parallel implementing probabilistic spreading algorithm using MapReduce programming mode

HE Dongzhi, ZHANG Jifeng, ZHAO Pengfei

Journal of Shandong University(Engineering Science). 2015, 45(5):  22-28.  doi:10.6040/j.issn.1672-3961.0.2014.367

Abstract ( 807 )   PDF (981KB) ( 439 )   Save

References | Related Articles | Metrics

In order to overcome the limitations of the serial probabilistic spreading algorithm in dealing with large-scale dataset, a parallelization of the algorithm was put forth by using MapReduce. The complex computing tasks were decomposed into a series of MapReduce job flow for distributed parallel processing on Hadoop. The input and output data of every step were stored in the Hadoop distributed file system. Hit ratio was used to compare the parallelizable probabilistic spreading algorithm versus the global ranking method performance. Speedups of the parallelizable algorithm were compared while the amount of data and the number of nodes was different. Experiment results showed that the probabilistic spreading algorithm based on MapReduce had good parallelism and had higher hit ratio than the global ranking method. Data scale that can be handled by the serial algorithm was expanded, and the operation speed of the algorithm was raised.

Ship detection in remote sensing image based on the fuzzy fusion of multi-channel Gabor filtering

XIAO Qiao, PEI Jihong, WANG Lixia, GONG Zhicheng

Journal of Shandong University(Engineering Science). 2015, 45(5):  29-35.  doi:10.6040/j.issn.1672-3961.3.2014.033

Abstract ( 859 )   PDF (2735KB) ( 486 )   Save

References | Related Articles | Metrics

A scheme to sea background suppressing was proposed for ships detection in optical remote sensing images based on the fuzzy fusion of multi-channel Gabor filtering. First, a multi-channel Gabor filter was designed to give output image group. Second, three filtering enhancement evaluations were defined to get the fuzzy evaluation matrix. Third, the fuzzy comprehensive evaluations were calculated and the significant images were selected from filtered output images. Finally, the weights of the significant images were determined and the fused image was given by using weighted sum of these significant images. Experimental results showed that the proposed ship detection algorithm based on fuzzy fusion of multi-channel Gabor filtering could efficiently improve the detection accuracy and significantly reduce false alarm rate.

Hierarchical cost sensitive decision tree and its application in the prediction of the mobile phone replacement

XIONG Bingyan, WANG Guoyin, DENG Weibin

Journal of Shandong University(Engineering Science). 2015, 45(5):  36-42.  doi:10.6040/j.issn.1672-3961.2.2015.190

Abstract ( 781 )   PDF (986KB) ( 281 )   Save

References | Related Articles | Metrics

In the data of mobile phone users, imbalance problem existed between the replacement users and non replacement users, however traditional date mining pursued the best overall accuracy which led the prediction accuracy of the replacement users overly low. In order to solve this problem, a method of predicting the users who replace phone was proposed based on hierarchical cost sensitive decision tree. The algorithm realized attributes reduction and calculated the importance of attributes by rough set, then a hierarchical structure was built by parting the attributes; finally a cost sensitive decision tree was regarded as the base classifier for the hierarchical structure, the decision tree was constructed with its splitting criterion which included gini index and misclassification cost. Three experiments were made for the users data which from a telecom operator, the results showed that the hierarchical cost sensitive decision tree achieved a better effect on the imbalance user data and balance user data which obtained by under sampling.

Two-degree-of-freedom Smith predictor control for cascade time delay process

ZHANG Jinggang, MA Wenting, ZHAO Zhicheng

Journal of Shandong University(Engineering Science). 2015, 45(5):  43-50.  doi:10.6040/j.issn.1672-3961.0.2014.355

Abstract ( 963 )   PDF (1170KB) ( 207 )   Save

References | Related Articles | Metrics

A novel two-degree-of-freedom(2DOF)Smith predictor control scheme for cascade time-delay process was proposed. The 2DOF Smith predictor control structure, which consist of setpoint tracking controller and disturbance rejection controller, was adopted in the main loop. And the internal model control structure was adopted in the secondary loop. The two controllers were respectively designed by an analytical method in the main loop. The setpoint tracking characteristics was decoupled from the disturbance rejection characteristics. Theoretical analysis and simulation results showed that the proposed control structure was simple, the controller parameters could be adjusted conveniently, and the system has a very good dynamic response performance and robustness.

Finite-time filtering for discrete-time linear impulsive singular systems

TONG Yunxu, LI Guihua, LIU Tingting, ZHU Yuqing

Journal of Shandong University(Engineering Science). 2015, 45(5):  51-57.  doi:10.6040/j.issn.1672-3961.0.2015.021

Abstract ( 907 )   PDF (854KB) ( 183 )   Save

References | Related Articles | Metrics

The finite-time filtering problem for discrete-time linear impulsive singular systems was investigated. Firstly, sufficient conditions for the filter error system which is finite-time stable and satisfies the performance requirements were given by using the linear matrix inequality method. Secondly, sufficient conditions for solving the finite-time filter problem of impulsive singular systems were presented, and the designing method of filter was provided, which generalizes the finite-time filtering theories of impulsive systems. Finally, a numerical example was given to demonstrate the feasibility of the conclusion.

Fractional order internal model control with fuzzy-tuning for DC speed regulating system

ZHAO Zhitao, ZHAO Zhicheng, WANG Huifang

Journal of Shandong University(Engineering Science). 2015, 45(5):  58-62.  doi:10.6040/j.issn.1672-3961.0.2015.004

Abstract ( 849 )   PDF (1854KB) ( 185 )   Save

References | Related Articles | Metrics

A fuzzy self-tuning fractional internal model controller was proposed for DC speed regulation system. Firstly, the theory of fractional calculus and internal model control(IMC)were combined to design a fractional internal model controller, and the parameters of controller could be tuned according to the phase margin and crossover frequency of the system. Then, based on analysis of the impact of the phase margin and crossover frequency on the system performance, the fuzzy controller was designed. So, the online self-tuning of the controller parameters was realized according to speed error and its change, and the system performance dependence on the phase margin and crossover frequency selection was overcome. The simulation and experimental results showed that the proposed controller could make the system have a better dynamic response characters, disturbance rejection performance and robustness against parameters perturbation of system.

Design and implementation of a coal loading control system with high accuracy and efficiency

ZHANG Wen, SONG Hongjun, LI Tianhua, LIU Yanfen

Journal of Shandong University(Engineering Science). 2015, 45(5):  63-69.  doi:10.6040/j.issn.1672-3961.0.2015.223

Abstract ( 695 )   PDF (6992KB) ( 292 )   Save

References | Related Articles | Metrics

A Kalman filtering based noise abating method about the impulse of coal loading was developed, an automatic control system of quantitative coaling using AB PLC was designed and implemented. A set of quantitative coaling control system consisted of blending logic control, performance monitoring,data integration was developed and deployed on real coal mine. After on spot test and data analysis, the load error of single train carriage was less than 0.1%, the entire train load error is less than 0.01%, the load capacity reaches 10 000 t/h. It greatly improves the work efficiency and automation level of the quantitative coal loading system.

Study on approaching excavation influence between expanding projects of expressway and high speed railway tunnel under operation state

SUN Keguo, QIU Wenge, WANG Zhongping, GAO Xiaoyan, GONG Lun, KONG Chao

Journal of Shandong University(Engineering Science). 2015, 45(5):  70-76.  doi:10.6040/j.issn.1672-3961.0.2014.369

Abstract ( 806 )   PDF (4185KB) ( 145 )   Save

References | Related Articles | Metrics

Relying on the expanding project of Lianhuo expressway Luosan segment crossing over the Wenxiang tunnel and utilizing the beam-slab protection scheme, the strength reduction method considering the reduction of both rock and liner was utilized to analyze the safety of maximum excavation and overload operation. The results showed that the maximum excavation was the most dangerous state and this scheme could guarantee the safety of both projects. This conclusion was dependable for constructing this adjacent project. With the help of quality testing for liner and the wireless monitoring method, this project was successfully constructed, which proved the result's reliability and could also provide technical reference for tunnel design and construction of similar adjacent projects.

New peak shear strength criterion with inclusion of shear action history

ZHANG Jianming, LIU Quansheng, TANG Zhicheng, ZHAN Ting, JIANG Yalong

Journal of Shandong University(Engineering Science). 2015, 45(5):  77-81.  doi:10.6040/j.issn.1672-3961.0.2015.072

Abstract ( 973 )   PDF (971KB) ( 275 )   Save

References | Related Articles | Metrics

Artificial joints were used for direct shear tests under four levels of normal stresses, i. e. 0.5,1.0, 2.0 and 3.0 MPa,to simulate the effect of shear deformation history. The results showed that peak strength decreased with the shear times increasing. In addition, the discrepancy between the result of direct shear test for the first time and for the second time was the most obvious. With shear times increasing, there were fewer changes of peak shear strength in following tests. Analyzing dilatancy characteristics and changes of peak shear strength under normal stresses, the peak shear strength criterion considering the effect of shear deformation history was established.

Construction and analysis on typical working cycle of wheel loader

MA Xiangming, SUN Xia, ZHANG Qiang

Journal of Shandong University(Engineering Science). 2015, 45(5):  82-87.  doi:10.6040/j.issn.1672-3961.0.2015.095

Abstract ( 907 )   PDF (1288KB) ( 323 )   Save

References | Related Articles | Metrics

In order to obtain more precise operating parameters of typical working cycle and simulate the fuel consumption, a constructing method of typical working cycle for wheel loader was proposed based on kinetic sequences and operating characteristics. Thirteen characteristic parameters, including speed-torque area etc., were defined to evaluate the variation of operating parameters during working process. Meanwhile, the kinetic sequences were classified into five clusters by the adoption of the Principal Component Analysis and fuzzy C-means algorithm. Afterwards, the sequences with minimum Euclidean distance in each cluster were selected to constitute the typical working cycle. The results showed that parameters of the test condition and the typical working cycle, such as acceleration and speed, had the similar distribution; fuel consumption based on the typical working cycle simulation was 0.176 L with an error of 4.7%, which was within the scope of the engineering allowable error. The proposed method could be effectively employed to process and analyze the test data of wheel loader and construct the typical working cycle, which was also greatly helpful to further study the fuel economy of loader working cycle.

The simulation of fluid dynamics and mass transfer in a double-loop WFGD

QIN Mingchen, DONG Yong, CUI Lin, SUI Hui, LIU Jinglong

Journal of Shandong University(Engineering Science). 2015, 45(5):  88-94.  doi:10.6040/j.issn.1672-3961.0.2015.159

Abstract ( 712 )   PDF (2336KB) ( 257 )   Save

References | Related Articles | Metrics

A gas-liquid hydrodynamic model was established by the Euler-Lagrange method to simulate a pilot-scale double-loop wet flue-gas desulfurization(WFGD)reactor. The SO₂ absorption mass-transfer model was primarily built on the dual-film theory, and the SO₂ absorption procedure was implemented in the commercial software FLUENT through a user-defined function(UDF). The relationship between main operation parameters and the SO₂ removal efficiency was analyzed, and the gas and liquid side mass transfer resistance was also discussed. Results showed that the desulfurization efficiency rised by increasing liquid-to-gas ratios and reduced with the improving of superficial gas velocity. Gas film resistance to liquid film resistance ratio was about 0.35 in the lower-loop, while 0.65 in the upper-loop. Studies on the pilot-scale double-loop WFGD tower showed that the simulated SO₂ removal efficiency agreed well with the experimental data, with an error of less than 5%. Therefore, the mathematical model can be used to optimize full-size double-loop WFGD structures, as well as to predict the SO₂ distribution and desulfurization efficiency.