Table of Content

25 June 2006, Volume 34 Issue 6

Previous Issue    Next Issue

Electric Engineering

Design and Implementation of Optimized Dispatching Software System for Unit Load in Thermal Power Plant

Ma Xiao-qian Wang Yi Liao Yan-fe

2006, 34(6):  112-116. 

Asbtract ( 279 )   PDF (357KB) ( 1682 )  

Related Articles | Metrics

In order to reduce the energy cost of thermal power plants,the architecture of a real.time optimized dispatching software system based on SIS(Supervisory Information System)platform is presented for unit load,in which the tight integration of SIS and MIS(Management Information System)and the auto-dispatching of the load are implemented.The real-time characteristic curves describing the energy cost of each unit are then simulated by the on-line collection and economic calculation of the operating data,with a mathematical model of load dispatching being set up.According to the proposed model,the optimized dispatching of the load is realized via the layered fuzzy genetic algorithm.Moreover,the corresponding dispatching software system is designed and implemented based on the Microsoft.NET Framework and the multi-layer Browser/Server concept.As compared with the traditional onduty dispatching mode,the proposed softwaFe helps decrease the power-supply coal cost rate by 0.5 g/(kW ·h), thus reducing the energy cost.

Application of Fuzzy Neural Network-Based Chaos Optimization Algorithm to Coal Blending

Liao Yan-fen Ma Xiao-qian

2006, 34(6):  117-121,126. 

Asbtract ( 338 )   PDF (352KB) ( 962 )  

Related Articles | Metrics

In view of the fuzzy and nonlinear characteristics of coal slagging process,a fuzzy neural network model is developed to evaluate the slagging characteristic of blended coal,with the slagging tendency index as the restriction condition of coal blending.Then,the coal blending in power plants is optimized by the chaos optimization algo.rithm with the ergodic advantage of chaotic motion in the solution space.The minimum price of blended coal is used as the optimization object and the restriction conditions are converted by means of the penalty function method.The results reveal that the combination of fuzzy neural network with the mutative.scale chaos algorithm shows good adaptability and practicability in the coal blending in p0wer plants.The proposed model can help acquire relative optimum results in a short time and is conducive to the on-line monitoring of real-time coal blending in power plants.

Application of Nonlinear M odels to Forecast Vibration Faults of Turbo-Generator Unit

Liang Ping Fan Li-li Long Xin-feng

2006, 34(6):  122-126. 

Asbtract ( 334 )   PDF (288KB) ( 927 )  

Related Articles | Metrics

In order to overcome the nonlinear change of the amplitude of turbo-generator unit in malfunction,a forecasting model combining GM (Gray Model)with ARMA (Auto Regression Moving Average),a forecasting model based on the fractal collage theorem and the fractal interpolation,and a forecasting model based on the leastsquare support vector machine are established,which are then respectively fitted with the daily average vibration peak-peak values of a 200 MW unit．At the same time,the forecasting performances of the three proposed models are analyzed and compared,with a suitable model obtained for the fault forecasting of turbo-generator units.

Electronics, Communication & Automation Technology

Improved SPIHT Algorithm Based on ROI Shape Estimation of M edical Image

xu Xiang-min Xing Xiao-fen Liu Wei Chen Xiao-chuan

2006, 34(6):  1-5. 

Asbtract ( 232 )   PDF (311KB) ( 872 )  

Related Articles | Metrics

In order to improve the transmission quality and coding eficiency of medical images,a novel bitplane lifting algorithm and an improved SPIHT(Set Partitioning in Hierarchical Trees)algorithm based on ROI(Region of Interest)shape estimation is proposed.By implementing the interlaced lifting of bitplane,the relative quality of the ROI and the background becomes adjustable without transmitting any mask information.For the bitplane after the lifting,the estimated mask inform ation can be obtained by transmitting the geometric parameter of the circum rule shape of ROI.According to the estimated mask inform ation,zero trees are determined be~rehand,so that the coding bits of the image can be saved.Experimental results show that,at the same truncation bit rate,both the ROI and the whole image obtained by the improved algorithm are of better image quality than those obtained by the traditional SPIHT algorithm,and that the lower the bit rate is,the more obvious the results are.

Image Registration Approach Based on Nearest-Point Pseudo Gravitation Fidd

Zhang Jian-wei Han Guo-qiang

2006, 34(6):  6-11. 

Asbtract ( 231 )   PDF (364KB) ( 388 )  

Related Articles | Metrics

In order to overcome the disadvantages of the image registration method based on Hausdorff distance field of image edges,an image registration approach based on the nearest-point pseudo gravitation field is presented,in which the edge points of the reference image are assumed to generate a nearest-point pseudo gravitation field,and the average force loaded on the edge points of the floating image in the gravitation field is adopted as the similarity function.This approach reduces the effect of excess part of the floating image,improves the precision of image regi-stration,and decreases the misregistration ratio.Thus,the registration of excess edges in the floating image is effectively solved.

Subcarrier Allocation Algorithm in M ultiuser OFDM Systems

Gong Han-dong Ye Wu Feng Sui-li Ke Feng

2006, 34(6):  12-16,54. 

Asbtract ( 290 )   PDF (334KB) ( 700 )  

Related Articles | Metrics

The application of adaptive allocation techniques to OFDM (Orthogonal Frequency Division Muhiple.xing)systems optimizes the use of system power.In this paper,a subcarrier allocation algorithm based on the re-quired data rates of the users and the characteristics of the channels is proposed by analyzing the power optimization problem.In the proposed algorithm,the subcarrier number finally allocated to each user does not need to be calcu-lated and the allocation process is divided into two steps,namely,the basic allocation and the residual allocation.In the basic allocation,the data of each user is transmitted in time ,and in the residual all ocation.the subcarriers are allocated to users according to the principle of reducing the achievab le minimum transmission power required by each user·Simulated results show that the proposed algorithm can efectively reduce the transmission power

Iterative Detection Algorithm for V-BLAST

Li Xiao-wei Wei Gang Chen Fang-jiong

2006, 34(6):  17-20. 

Asbtract ( 274 )   PDF (249KB) ( 548 )  

Related Articles | Metrics

The optimal maximum likelihood(ML)detection of vertical-bell laboratory layered space-time codes (V-BLAST)is dificult to implement and the performances of conventional detection algorithms are limited by error propagation.In order to solve these problems,an iterative detection(ID)method for V-BLAST codes is proposed,in which the soft inform ation of the signals received by multiple antennas is iterated to eliminate the interference.By the proposed method,the effect caused by the error propagation is efficaciously mitigated.Simulated resuhs indicate that the ID method is of quicker convergence speed and better detection perform ance,as compared with the conventional ZF-DFE algorithm.Moreover.it is of a detection perform ance similar to the ML method in low SNR (Signal Noise Ratio)region.

Performance of MIMO System in Spatially Correlated Fast Rician Fading Channel

Lai Guo-ting Yin Jun--xun Yu Hua-wen

2006, 34(6):  21-24. 

Asbtract ( 249 )   PDF (220KB) ( 994 )  

Related Articles | Metrics

By considering the advantages of MIMO(Muhi-Input Muhi-Output)system in transmitting and receiving diversity as well as the universality of spatially correlated fast fading Rician channels,a close-form solution to the upper bound of the average pairwise eiTor probability of MIMO system with spatially corelated fast Rician fading is derived based on the multivariate statistic theory.It is found that the system perform an ce tends to decrease when the
spatial corelation of channels increases or the Rician factor decreases.The corectness of the theory derivation is fi-nally verified by simulation.

RFID Tag Antennas Based on Hilbert Fractal Structure

2006, 34(6):  25-28. 

Asbtract ( 294 )   PDF (213KB) ( 953 )  

Related Articles | Metrics

A miniaturization structure of RFID (Radio Frequency of Identification)tag antennas based on Hilbert fractal structure is proposed.Then,the resonant frequencies ,radiation patterns and efficiencies of the tag antennas are simulated by means of MoM (Method of Moment).and a one.dimension tag antenna with Hilbert fractal struc ture is manufactured for test.The simulation and test results show that the space.filling feature of Hilbert fractal
structure antenna can be effectively transformed into the size.reducing feature.and that the RFID tag an tenna with one-dimension Hilbert fractal structure iS of higher emciency

Output Tracking Control of M obile M anipulators Based on Dynamical Sliding-Mode Control

2006, 34(6):  29-33. 

Asbtract ( 286 )   PDF (244KB) ( 569 )  

Related Articles | Metrics

A dynamical sliding-mode controller is devised to track the output of mobile manipulators.During the in-vestigation,a reduced dynamic model considering the dynamics of the driving motor is first developed for mobile manipulators.Then,the system is decomposed into four lower-dimension subsystems by means of diffeomorphism and nonlinear input transformation.Moreover,a design method of the dynamical sliding-mode controller applied to the output tracking of mobile manipulators is proposed.The simulation results indicate that the dynamical sliding mode controller can not only track the given trajectory correctly but also reduce the chattering of sliding-mode con-trol systems considerably.

Design and Application of On-line Neural Network-Based Adaptive Controller

Tang Chao-ying Wang Biao

2006, 34(6):  34-38. 

Asbtract ( 274 )   PDF (250KB) ( 556 )  

Related Articles | Metrics

An on-line neural network.based adaptive controller is presented to solve the problem of weak adaptability of existing linear control systems.The network weights are regulated according to the error between the reference model and the plant.and the structural uncertainties and unknown disturbances are counteracted on line by the neural network.The proposed controller is then designed and simulated with the attitude control of a spacecraft as an example.The results indicate that the uncertainties are tracked and compensated on line by the neural network with high accuracy and the dynamic performan ce of the linear control system is considerably improved.

Optimal Feedback Scheduling Based on Anytime Algorithm for Model Predictive Controllers

Zhou Ping-fang Xie Jian-ying Deng Xiao-long

2006, 34(6):  39-43. 

Asbtract ( 266 )   PDF (319KB) ( 866 )  

Related Articles | Metrics

A model predictive controller can be implemented as a MPC(Model Predictive Control)task with the characteristics of the Anytime algorithm ,which allows the computation time to be traded for control performance.This paper presents an optimal feedback scheduling(FS·CBS)algorithm for a set of MPC tasks to maximize the global control performance subjected to limited processor time.Each MPC task is assigned with a CBS(Constant Bandwidth Server),whose reserved processor time is dynamically adjusted.The constraints in the FS-CBS algo-rithm guarantee the schedulability of the total task set and the stability of each component．Simulated results indi-cate that the optimal scheduling algorithm is robust against the variation of execution time of MPC tasks at runtime,and that it performs much better than the basic CBS algorithm.

Absolute Stability of Lurie Uncertain Systems with Multiple Time-Varying Delays

Ma Xin-Jun Xu Bu-gong Xiang Shao-hua Huang De-xian

2006, 34(6):  44-48. 

Asbtract ( 278 )   PDF (225KB) ( 421 )  

Related Articles | Metrics

Based on the Lyapunov-Krasovskii function and by means of the matrix inequality technique and the non-linear dealing method,Lurie uncertain systems with multiple time-varying delays are investigated.Two new suffi-cient conditions of absolute stability described by linear matrix inequalities are acquired and they are characterized by norm bounded uncertainties and time-varying delays.Calculation examples reveal that the proposed method is less conservative and more efective than the existing ones in the literature.

QLMBP：A Quick BP Neural Network Algorithm

Li Jiong-cheng Huang Han-xiong

2006, 34(6):  49-54. 

Asbtract ( 273 )   PDF (342KB) ( 621 )  

Related Articles | Metrics

The improved version of BP(Back Propagation)algorithm with the fastest convergence speed,LMBP (Levenberg-Marquardt BP)algorithm,is investigated,finding out the bottlenecks of the convergence speed,that is,the initialization of iteration controlling parameters has a great influence on the iterated number.the calculation of the inverse matrix involved in each iteration is the most time-consuming ,and it will take long to carry out a certain interation if the sum of squared el'TOrs in each interation is not decreased.To solve these problems,LU(Lower-Upper)decomposition is employed to avoid the time-consuming calculation of inverse matrix.and the one-di-mension searching is adopted to accelerate the decrease of the object function.Thus,a quick BP neurM network al-gorithm named QLMBP(Quick LMBP)is proposed.The proposed QLMBP algorithm is independent on the itera tion controlling parameters and its convergence speed is about 100 times that of the LMBP algorithm convergence speed.

Taxation Forecasting M odel Based on BP Neural Network

2006, 34(6):  55-58. 

Asbtract ( 304 )   PDF (242KB) ( 1004 )  

Related Articles | Metrics

In this paper,the BP(Back Propagation)neural network theory is applied to forecast taxation after ana’ lyzing the major factors affecting the tax.During the investigation,the original data are preprocessed to meet the re quirements of the study in BP neural network,and a taxation forecasting model based on BP neural network is es tablished.The proposed model is then verified by using actual data and is compared with the traditional statistic model.It is concluded that the proposed model is of hi gh precision and great applicability.

Stochastic-Volatility Pricing Model with Jump and Feedback for Securitie

Sun You-fa Deng Fei-qi

2006, 34(6):  59-63. 

Asbtract ( 314 )   PDF (300KB) ( 428 )  

Related Articles | Metrics

Until now there exists no universally accepted security pricing model that can exactly describe the beha-viors of real stock prices.In this paper,a new security pricing model-stochastic-volatility pricing model with jump and feedback- is proposed by considering the fact neglected by the traditional pricing models that important events frequently happen in the financial market.The interaction between investors and security prices is also taken into
account.Theoretical analyses,numerical simulations and practical applications all indicate that the proposed model can simulate the complex behaviors of real security prices better than the traditional models and has the merits of high precision and efficiency in prediction.

On-line M ulti-Output Support Vector Regression and Its Application to Investment Decision

Hu Gen-sheng Deng Fei-qi

2006, 34(6):  64-68. 

Asbtract ( 350 )   PDF (272KB) ( 770 )  

Related Articles | Metrics

An on-line multi output support vector machine regression algorithm is proposed in this paper.By using the gradient descent algorithm to minimize the instantaneous regularized risk of prediction results,the iterative for mulae of the weight coefficients and the bias of the regression function are obtained.Thus,the on-line multi-output regression prediction can be implemented for new arriving samples.The proposed algorithm is then applied to the investment decision to predict the optimal portfolio on line.Simulation results show that the proposed algorithm is easy to carry out because of its simple computation and small workload.

Computer Science & Technology

Algorithm of Classification Rules Based on Co-Evolution Computation

Min Hua-qing Lu Yan-sheng Jiang Xiao-yu

2006, 34(6):  69-73. 

Asbtract ( 296 )   PDF (371KB) ( 444 )  

Related Articles | Metrics

In order to improve the accuracy of classified mining,the ID3,C4.5 and EC(Evolution Computation) algorithms are analyzed,and two co-evolution populations are designed to respectively describe the attribute set an the classification rule set.The algorithm of the classification rules based on the co-evolution computation f CRCEC) and its fitness function are then proposed.Moreover,a comparison among EC,ID3,C4.5 and CRCEC algorithms
is carried out using four datasets of University of California.Irvine.The results show that the proposed CRCEC al.gorithm is of high accuracy and helps obtain rules that are simple and easy to understand.The proposed algorithm is finally applied to the predictive system of highway charge as an application example.

Determination of Related Web Queries Using Support Vector Regression

Wang Ji-min Peng Bo Meng Tao

2006, 34(6):  74-78,94. 

Asbtract ( 258 )   PDF (430KB) ( 609 )  

Related Articles | Metrics

When a user submits a Web query to a search engine,it is helpful for the user to modify the query and find the needed information if the system returns a list of related Web queries.This paper presents a new determ ina-tion method of related Web queries using support vector regression.In this method,five quantified indexes of a candidate query are extracted from the log files,including the submitted number of the candidate query ,the total numbers of submitting the candidate query and hitting the returned resuh,the number of common terms and the number of hitting common URL(Uniform Resource Locator)between the candidate query and the given query.The obtained candidate queries are then ranked based on support vector regression models learned from parts of human.1abeled training data.The related Web queries are finally determ ined according to the relevance.Experimental re-suits show that the proposed method is of high prediction precision.

Anyeast Routing Algorithm Based on Expanding Method

Li Ling-zhi Zheng Hong-yuan Wu Qing-feng Ding Qiu-lin

2006, 34(6):  79-83. 

Asbtract ( 306 )   PDF (343KB) ( 536 )  

Related Articles | Metrics

As anycast addresses are of insufficient quantity,high employing frequency and multiple hosts,an anycast routing algorithm based on the expanding method is proposed,in which the router joins the group member domain according to the computed metric that integrates the expand message sent by the anyeast server and the ca-pability of the network and creates the corresponding entry in its route table.Then,the request with this anycast address as the destination is forwarded to the group member through a direct mode or a tunneling mode.The pro-posed anycast routing algorithm was simulated by means of the network simulation software NS-2 and the results show that the proposed algorithm helps reduce the time delay of transmission and can thus improve the expansibility of services.

Admission Control Based on Centralized Bandwidth Broker for Diferentiated Service Networks

Huang Cheng-bo Zhang Ling Zhou Jie

2006, 34(6):  84-88. 

Asbtract ( 285 )   PDF (314KB) ( 581 )  

Related Articles | Metrics

The admission control schemes based on the centralized BB(Bandwidth Broker)are effective approa-ches to the providing of End-to-End QoS(Quality of Service)guarantee for differentiated service networks.Unfortunately,the existing schemes cannot effectively support the End-to-End QoS signaling and cannot solve such problems as the state management and the congestion control when rerouting occurs.To solve these problems,an admis-sion control scheme based on the centralized BB is designed.The architecture of the QoS state information base and the algorithms for admission control and state management are then given.and the policies for the state management and the congestion control when rerouting occurs are presented.Th e effectiveness of the proposed scheme is finally demonstrated by means of the network simulation software NS-2.

Congestion Control Scheme Based on Network Measurement and Fuzzy Logic Control Technology

Song Li-hua Wang Hai-tao Chen Ming

2006, 34(6):  89-94. 

Asbtract ( 339 )   PDF (449KB) ( 792 )  

Related Articles | Metrics

A network-measurement-based congestion control scheme is proposed to avoid the limitations resuhing from the AQM (Active Queue Management)mechanism and improve the TCP performance by feeding back timely and correct information on network states to end systems.In the proposed scheme,the measurement facilities dis.tributed in networks ale exploited to acquire the perform ance of network backbone,an d according to the resulting data a fuzzy logic controller is used to help end systems determ ine the appropriate FAST control parameters. Simula-tion experiments indicate that the proposed scheme can support extremely heavy load,achieve high throughput,and steady the queuing delay.Moreover,it behaves more stably and fairly than the AQM scheme in high-speed networks,and is more suitable for P2P like large-volume long.lived flows rather than Web.like bursty traffic.

Multi-Secret Sharing Scheme with General Access Structures

Li Hui-xian Cheng Chun-tian Pang Liao-jun

2006, 34(6):  95-98. 

Asbtract ( 347 )   PDF (269KB) ( 468 )  

Related Articles | Metrics

In order to widen the application of multi-secret sharing schemes,a multi-secret sharing scheme with general access structures was proposed based on Shamir’s threshold secret sharing scheme,in which multiple secrets are shared in each sharing session,and the secret shadow of each participant is reused,with a length as long as that of one shared secret.All these are different from the existing schemes.Analytical results show that,as COBpared with the existing schemes,the proposed scheme reduces the computational complexity of secret distribution and secret reconstruction algorithms and that it implements the sharing of multiple secrets,thus improving the sys tem performance.

Knowledge Propagation Model in Complex Networks

Li Jin-hua Sun Dong-chuan

2006, 34(6):  99-102. 

Asbtract ( 359 )   PDF (260KB) ( 1434 )  

Related Articles | Metrics

According to the propagation characteristics of knowledge in the collaborative networks of knowledge,a knowledge propagation model is proposed based on the complex network theory.In the proposed model,the Cobb-Dauglas production function is introduced to describe the increase of knowledge caused by knowledge diffusion,a kind of collaborative production of knowledge.The knowledge propagation with or without the self-improvement of agents are then considered,with the results indicating that,when other conditions remain constant,the diffusion speed and the distribution uniformity of knowledge in networks increase with the stochastic degree of networks.

Implementation of Adaptive DPM Algorithm Based on Stochastic M odel in Linux System

Liu Fa-gui Chai Yang-yang Liu Yong Xi Jian-qing

2006, 34(6):  103-107. 

Asbtract ( 287 )   PDF (370KB) ( 836 )  

Related Articles | Metrics

In view of the rapid increase in the power consumption of electronic equipment at present,the DPM (Dynamic Power Management)in Linux system is investigated.An adaptive DPM algorithm based on the stochastic model is introduced and the corresponding implementation in Linux system is then presented.In the implementation of this proposed algorithm.the code of the observer in the kemel is used to mentor the use of system resources;PCx
is adopted to carry out the linear programming asking and solving in the optimization process;and the controller sends out hardware order to directly control the hardware.Experimental results indicate that the proposed DPM al-gorithm can effectively reduce the energy cost of the system without great modification of the Linux kemel,and that the effects resulting from the algorithm can almost be neglected.

Extended Subjective Logic Based on Dirichlet Distribution

Su Jin-dian Guo He-qing Liu Miao

2006, 34(6):  108-111,116. 

Asbtract ( 273 )   PDF (295KB) ( 640 )  

Related Articles | Metrics

In order to solve the problems existing in the binary logic-based subjective logic,this paper proposes an extended subjective logic based on the Dirichlet distribution.By using the ternary logic instead of the binary logic,the extended subjective logic redefines the mapping relationship and functions between the evidence space and the opinion space,and presents a new extended consensus rule.Moreover.it can model and deal with the subjectivity and the uncertainty of trust well because it retains the advantages of combining statistic inference and probability theory in subjective logic and takes into consideration the uncertain results of an event.Results 0f sample analvses show that the proposed extended subjective logic is reasonably supported by practical interpretations as well as by theoretical foundations.

Contents

2006, 34(6):  0. 

Asbtract ( 59 )  

Related Articles | Metrics