Table of Content

30 August 2011, Volume 45 Issue 08 Previous Issue    Next Issue

Automation Technique, Computer Technology

Select

Intelligent Control of Fused Magnesium Furnaces Based on SPSA WU Zhi-Wei, WU Yong-Jian, CHAI Tian-You 2011, 45 (08):  1095-1100.  Abstract ( 5565 )   Save The adjustment of electrode currents for smelting process of fused magnesium furnace depends manual control mainly and the control effect is very poor, the currents fluctuation is very huge which leads to the instability of the product quality and great energy waste. As such, an intelligent control strategy was developed for fused magnesium furnace. The neural network controller based on SPSA algorithm and another controller based on rulebased reasoning were designed respectively according to different states. The electrode currents can be controlled automatically and the energy consumption is reduced based on the intelligent control strategy. The proposed control system  is applied in a magnesium plant and the result shows its validity and effectiveness.   References | Related Articles | Metrics

Select

Estimation of Performance Benchmark for a Class of Nonlinear Control Systems ZHANG Zhi, HU Li-Sheng 2011, 45 (08):  1101-1104.  Abstract ( 3890 )   Save  An estimation problem of the performance benchmark is considered when the control loop is intervened by the complex nonlinearity. The nonlinearity includes control valve stiction and switching action of the protection valve. The proposed method is based on the threshold nonlinear autoregressive exogenous (TNARX) model. The input and output routine operating data are partitioned if they reaches the certain thresholds. Modeling each partitioned regime as a NARX model, then the prediction error variance as the performance benchmark can be obtained by the principle of pooled variance. By comparison with polynomial autoregressive, polynomial autoregressive exogenous and selfexciting threshold autoregressive models, a simulation example shows the effectiveness and advantage of the proposed method. References | Related Articles | Metrics

Select

The Optimized Scheduling for IronMaking Bulk Ore Blending Process Based on Improved Ant Colony Optimization HU Kai-Lin, LI Ping 2011, 45 (08):  1105-112.  Abstract ( 4465 )   Save A discrete scheduling model for the ironmaking bulk ore blending process was proposed by converting the process into a two stage assembly flow shop with sequencedepended setup times and limited intermediate buffer. To find the optimal solutions of this problem, an improved ant colony optimization (ACO) algorithm was also proposed, in which a tabu timer is set for each ant to correct the feasible path set. And the efficiency and solution quality of this algorithm for solving this particular scheduling problem is convinced by the computational experiments. References | Related Articles | Metrics

Select

A DataDriven Approach for Model Predictive Control Performance Monitoring ZHANG Guang-Ming, LI Ning, LI Shao-Yuan 2011, 45 (08):  1113-1118.  Abstract ( 4247 )   Save A datadriven approach for model predictive control performance monitoring was proposed. An overall performance index based on Mahalanobis distance was introduced with its benchmark deduced to achieve higher monitoring performance. To identify the root cause of performance degradation, Mahalanobis distance based performance diagnosis method was proposed. Performance signatures are extracted from both principal component and residual subspace, and a classifier is constructed to identify four common performance degradation patterns. The effectiveness of the proposed method was demonstrated in a case study of the WoodBerry distillation system. References | Related Articles | Metrics

Select

Operating Parameters Setting Method of Sintering Waste Heat Recovery Process Based on Heat Production Optimization XIONG Yong-Hua, JIANG Yun-Tao, WU Min, LAI Xu-Zhi 2011, 45 (08):  1119-1124.  Abstract ( 3375 )   Save  In order to improve the efficiency of waste heat recovery system within sintering process, this paper introduced the index of boiler’s effective heat production to evaluate the overall efficiency of waste heat power generation system, and proposed an operating parameters optimization method based on this index. First, the BP neural network is applied to predict the index, so that the optimization model for boiler’s effective heat production is constructed. Based on this model, an improved particle swarm optimization is developed, in order to obtain the optimal values of middlepressure steam’s temperature and coolmachine speed, which result in the maximal boiler’s effective heat production. The simulation results show that the above method is able to improve averagely the boiler’s effective heat production of waste heat recovery system about 5%, and increase the efficiency of waste heat power generation. References | Related Articles | Metrics

Select

A Predictive Model of Short-Term Wind Speed Based on Improved Least Squares Support Vector Machine Algorithm ZHANG Guang-Ming, YUAN Yu-Hao, GONG Song-Jian 2011, 45 (08):  1125-1129.  Abstract ( 5104 )   Save In order to improve the forecast precision, an improved wind speed forecasting algorithm was discussed. The new method has modified extreme points and processed offset of predicting data, considering with the extreme points of the change in wind speed affecting the prediction error and the delay of prediction curve compared with actual wind speed. The forecasting model has better prediction accuracy and better computing speed to predict wind speed for the next one hour, compared with the wind speed model based on least squares support vector machine optimized by particle swarm optimization algorithm(PSO-LS-SVM), least squares support vector machine (LS-SVM) and back propagation (BP) neural network. The simulation results show that the improved least squares support vector machine is an effective method for short-term wind forecasting. References | Related Articles | Metrics

Select

Salient Building Region Detection for Scene Classification CHEN Shuo, YU Xiao-Sheng, WU Cheng-Dong, CHEN Dong-Yue 2011, 45 (08):  1130-1135.  Abstract ( 4893 )   Save In order to achieve outdoor scene classification in urban areas quickly, an effective method for detecting salient building region based on visual attention selection mechanism was proposed. Firstly, eightdirectional Gabor filter is used to obtain horizontal-vertical enhanced image. Secondly, salient building candidate regions are extracted through mapping from salient map which is constructed by phase Fourier tranform to the enhanced image. Finally, an orientation histogram discrimination algorithm is used to eliminate interference target, and salient building region detection is completed by CV model segmentation. The method extracts the building regional information which can represent scene based on PFT and orientation features, it effectively improves the real-time of algorithm and the capability of scene analysis, and implements the accurate detection of region in pixel level. The results of experiments with the university of Southern California outdoor scene database show that the method performs well in the detection result, computational speed and robustness. References | Related Articles | Metrics

Select

Robust SoftSensing of Slurry pH Using LSSVR for Mineral Flotation Process REN Hui-Feng, YANG Chun-Hua, ZHOU Xuan, GUI Wei-Hua, YAN Feng 2011, 45 (08):  1136-1139.  Abstract ( 3354 )   Save  Considering the poor stability of detectors and serious manual detection timedelay, a novel soft sensor was proposed based on least squares support vector regression (LSSVR) with sparsity using image features as instrumental variable. Firstly, multiple kernels were combined and the kernel matrix was reduced according to an improved minus cluster algorithm. Then the partial least squares regression was used to improve the robustness and precision of the soft sensor. The experiment verified the presented model which performs high precision and good reliability compared with standard LSSVR, weighted LSSVR and multiplekernel LSSVR. References | Related Articles | Metrics

Select

Fast Leave-One-Out Cross-Validation Algorithm for Extreme Learning Machine LIU Xue-Yi-a, LI Ping-a, b , GAO Chuan-Hou-c 2011, 45 (08):  1140-1145.  Abstract ( 5556 )   Save  Leaveoneout crossvalidation has proved to be near capable of giving the unbiased estimation of the generalization performance of statistical models, and thus can provide a reliable criterion for model selection and comparison. For this reason, the current paper presented a fast leaveoneout crossvalidation algorithm in the framework of extreme learning machines (ELMs) with respect to both regression and classification problems, which can avoid training explicitly and just has the complexity of O(N) for a data set with N points. The validity of the algorithm is also strictly proved. The simulations conducted on the artificial and realworld problems show the effectiveness and efficiency of the proposed algorithm. References | Related Articles | Metrics

Select

Robust Control Based on LMI for Satellite Attitude Control Systems HAN Xiao-Dong-1, BAO Cui-Mei-2, ZHANG Deng-Feng-3 2011, 45 (08):  1146-1150.  Abstract ( 3850 )   Save Considering the actuator faults, the solution of this problem was obtained for an Hinfinity statefeedback controller with circular disk pole constraints. Based on linear matrix inequality, the consistence on circular disk pole index and Hinfinity index was analyzed, and sufficient conditions for the existence of robust Hinfinity faulttolerant controller were given. The simulation of applying in a satellite attitude control system with actuator faults show the validity of the proposed method. And the results shows that this method has better robustness for satellites attitude control actuator under the condition of performance degradation or fault state，which is capable of meeting the attitude control requirement. References | Related Articles | Metrics

Select

Global Asymptotic Stability Criteria for Stochastic Neutral-Type Delayed Neural Networks LIU Guo-Quan-1, YANG Xian-Yi-1, CHAI Yi-1, FU Wei-1, FENG Wei-2, SHI Guang-1 2011, 45 (08):  1151-1157.  Abstract ( 4164 )   Save  The problem of global asymptotic stability analysis for a class of stochastic neutral-type neural networks with variable delays was investigated. By constructing an appropriate Lyapunov functional combined with inequality technique，sufficient stability criteria dependent on the size of the time delay were obtained in terms of linear matrix inequality (LMI). A class of neutral-type networks considered in the paper consists of distributed delays, stochastic perturbation and activation functions which are assumed to be constants (positive, negative, and zero), so these proposed criteria are universal and representative. The proposed stability results are more universal and less conservative than the existing ones.  Finally, two numerical examples were given to demonstrate the effectiveness of the proposed results. References | Related Articles | Metrics

Select

The Optimization Method and Simulating System of Distillation Column with Side Reactors for Benzene Chloride Production BAO Cui-Mei-a, b , TANG Ji-Hai-a, QIAO Xu-a, DING Liang-Hui-a, CUI Mi-Fen-a 2011, 45 (08):  1157-1161.  Abstract ( 3870 )   Save For the optimizing design problem of matching condition with reactive capability and separation capability for the distillation column coupled with side reactors (SRC), an independent reaction amount in simulating the SRC process was proposed to decouple the reaction kinetics from the mathematic models. A mixedinteger nonlinear programming (MINLP) optimization design method of the maximum reaction capability under the minimum product cost condition was discussed, and the simulating system software was programmed using MATLAB. The method was applied to the chloride benzene reactive distillation, the optimum matching condition of distillation and reaction capability was gotten which proves the integrated optimization design method is effective. References | Related Articles | Metrics

Select

Multi-objective Optimization of Fused Magnesium Enterprises Production Plan KONG Wei-Jian, CHAI Tian-You, DING Jin-Liang 2011, 45 (08):  1162-1166.  Abstract ( 3369 )   Save  Most of fused magnesium enterprises are productivitylimited factories, which will lose the orders that surpass their capacity in midseason and will overstock in off-season. Therefore, it was proposed to multistep forward predict requirements based on seasonal ARIMA model and to build the multi-objective optimization model of sales and stocks. A backward derivation method was designed to establish production plan, which can get the satisfying results. Aiming at prediction error, the real sales data are fed back to revise prediction results and adjust the optimization model to tune later production plan. The practical application proves the effects of the proposed approach.   References | Related Articles | Metrics

Select

Performance Analysis and Online Fuzzy Self-Tuning of RTD-A Controller’s Parameters CHU Ji-Zheng, DU Bin, CHEN Juan 2011, 45 (08):  1167-1171.  Abstract ( 3578 )   Save To reduce the difficulty of the RDT-A controller’s parameters tuning and enhance its utility, the effect of its four parameters（θR， θT， θD， θA）on system performance was analyzed. The result shows that it has rare influence on the system performance when θA and θT are set value, and the model mismatch and disturbances are improved after tuning of θR and θD. On this basis, an online autotuning method based on fuzzy logic was proposed for the parameters of the RTD-A controller. Fuzzy control rules of controller parameters were firstly established in this method, and then fuzzy reasoning was carried out according to the fuzzy control rules, and parameters θD and θA of fuzzy self-turning online were implemented. The simulated results show that the controller’s performance is evidently improved with the proposed parameter autotuning method, and compared with manual operation, the controller tuning is greatly simplified. References | Related Articles | Metrics

Select

A Multi-model Modeling Method Based on Supervised Affinity Propagation Clustering Algorithm DENG Wei-Wei-1, YANG Hui-Zhong-1, 2 2011, 45 (08):  1172-1175.  Abstract ( 4288 )   Save The traditional multimodel modeling method has big error because it does not consider the output error of the model during the clustering process, A multi-model modeling method based on supervised affinity propagation clustering algorithm was proposed. The principle is that the initial clusters are first obtained by the affinity propagation clustering algorithm, and then the clusters are adjusted cycledly in accordance with the output errors until the clusters do not change. Finally, the accurate clusters are got, and the sub-models are respectively built by least squares support vector machine so as to estimate the output. The method is used for the softsensor model to estimate the content of acetone at the outlet of a reaction vessel in a bisphenol A production process. The simulation results show that the method can get better modeling results than the traditional multi-model modeling method. References | Related Articles | Metrics

Select

Positive Real Control for Continuous System with Multiple State-Delays MA Jing-Wei-1, ZHAO Hong-Chao-1, ZHANG Xiang-Long-2 2011, 45 (08):  1176-1180.  Abstract ( 4088 )   Save The conditions of multiple state-delays and actuator failures weren’t simultaneously considered in former research. The continuous system with multiple state-delays was researched. The sufficient condition that this class of system is asymptotically stable and extendedly strictly positive real was deduced by theoretical analysis. On the basis of it, this class of system with uncertainties and actuator failures was researched, and a positive real controller was designed. It ensured that the close-loop system is asymptotically stable and extendedly strictly positive real against facultative uncertainties and actuator failures. Simultaneity, an example was analyzed to show that the proposed design approach of positive real controller is feasible and effective.   References | Related Articles | Metrics

Select

xPC Real-Time Simulation System Based on Multi-host PC Framework WANG Peng, LI Ping-Kang, YU Jun, YIN Chang-Zhi 2011, 45 (08):  1181-1185.  Abstract ( 4714 )   Save The approaches to the real-time control based on Matlab were discussed, a novel real-time simulation system framework based on “Multi-host PC” mode was proposed. Compared to traditional system, this system greatly improves the data processing capabilities and expands the scope of the hardware. Based on this new structure, a real-time target-tracking system of binocular vision two-wheeled robot was designed. Meanwhile, combined with M file-S functiondriven GUI real-time monitoring interface, the dynamic monitoring capabilities of system were enhanced. Finally, it was verified by test that the proposed multi-host PC framework can be effectively applied to real-time simulation system development. References | Related Articles | Metrics

Select

Equalization Control Method Design for Power Batteries in an Electric Vehicle YANG Chun-Lei, LIU Zhi-Yuan 2011, 45 (08):  1186-1190.  Abstract ( 3179 )   Save Considering the problem of the power batteries equalization control in an electric vehicle, a kind of equalization control method which is based on voltage transformer and constant current circuit was introduced. This method meets the requirement of equalization, and avoids the neglect effect of plus voltage. And then, the whole equalization structure of equalization was also designed, and the detailed design method of voltage transformer was introduced. Furthermore, considering the nonlinear character in the battery charging processing, a kind of close loop feedback control method which is easy to realize and meet the requirement of equalization was given. The experimental result proves that the designed method can equalize the batteries. References | Related Articles | Metrics

Select

A New AQM Algorithm Based on H∞ Control HOU Ping-1, 2 , WANG Zhi-Quan-1 2011, 45 (08):  1191-1195.  Abstract ( 4085 )   Save  An active queue management algorithm was presented for network congestion, which is based on uncertain parameters and variable link bandwidth. The available link bandwidth is modeled as a nominal constant value, which is known to the link, plus a time-variant disturbance, which is unknown. The model of TCP/AQM including the state and the input delay was presented in state variables. Then, the network congestion problem is solved by using the time-domain H∞ control approach. The NS2 simulation results show that this algorithm is capable of achieving the desired queue length fast under the conditions of variable RTT and sudden traffic flow. The performances of the proposed algorithm increase greatly compared with the existing algorithm. References | Related Articles | Metrics

Select

Satisfactory Control for a Class of Uncertain Switched Linear Systems DONG Xue-Ping, SU Yu-Tian 2011, 45 (08):  1196-1201.  Abstract ( 3548 )   Save The problem of satisfactory control for a class of uncertain switched linear systems was studied. The state feedback controller with closed-loop H infinity index and every subsystem with circular pole index, feedback gain and steady variance was designed. A sufficient condition for the switched system with above four indices was obtained. Meanwhile, by linear matrix inequalities (LMI) approach, the design of state feedback controllers with above four indices was transformed into the feasible problem of some certain LMI. The state feedback controllers can be efficiently solved by means of Matlab LMI toolbox. Finally, a numerical example shows the validity of the proposed design method. References | Related Articles | Metrics

Select

Nonlinear Dynamic Fault Detection Method Based on Isometric Mapping ZHANG Ni, TIAN Xue-Min 2011, 45 (08):  1202-1206.  Abstract ( 4207 )   Save The data collected from chemical process are strongly nonlinear and dynamic related. To solve this problem, a nonlinear dynamic fault detection method using dynamic isometric mapping (DISOMAP) manifold learning was proposed. It first extracts submanifold feature from original data set with adaptive neighbor parameters, which preserves geometric structure. Then linear regression projection mapping which maps the original high dimension space to a low dimension embedding space is used. Finally, T2 and SPE statistics are constructed in the process monitoring application. The simulation results of Tennessee Eastman process show that DISOMAP-based method is more effective than KPCA (kernel principal component analysis) for process monitoring and fault detection. References | Related Articles | Metrics

Select

Design and Implementation of Multi-function Vehicle Bus Card Based on USB Interface LIU Quan-Li, SONG Jian-Jun, WANG Wei 2011, 45 (08):  1207-1210.  Abstract ( 3466 )   Save The architecture of the MVB card was analyzed and the logic design of FPGA was described by VHDL language. A USB bus was used as an interface to realize the high speed data transmission between the FPGA and the computer, and the communication between the computer and the MVB card was realized by this way. The hardware design of the USB interface, the software design of the FPGA and the computer were described. Now, this MVB card is successfully used in underground train communication control systems. References | Related Articles | Metrics

General Industrial Technology

Select

A Copper Compositions Soft Sensor Using Color Vision and LSSVR ZHANG Hong-Wei, SONG Zhi-Huan 2011, 45 (08):  1211-1215.  Abstract ( 3016 )   Save This paper presented a new copper compositions soft sensor using color images of copper and LSSVR（least square support vector regression） modeling, as opposed to its delayed measurement by hardware measuring device. The soft sensor is based on a color vision system which includes computer and a color camera. First, a highperformance color chargecoupled device camera captures color from copper samples and transmits images to the computer. Then RGB, hue and color vector angle are used to characterize copper samples’ color for the comparative purpose. Three different LSSVR based soft sensor were developed using RGB, hue and color vector angle respectively. Finally, the real time estimation obtained by the soft sensor achieves superior performance. The experiments on test copper samples show that the proposed method is feasible and efficient. References | Related Articles | Metrics

Select

A Real-Time Lifetime Prediction Method Based on Wavelet Support Vector Regression Optimized by GA HU You-Tao, HU Chang-Hua 2011, 45 (08):  1216-1220.  Abstract ( 12566 )   Save Aiming at the fact that the present real-time lifetime prediction methods do not take full advantage of the same kind of products’ performance degradation data, as viewed from the comparability of degradation paths, a real-time lifetime prediction method was proposed on the basis of wavelet support vector regression (WSVR) optimized by genetic algorithm (GA). Firstly, GA-WSVR is employed to build the same kind of products’ performance degradation path models. Then the specific individual’s historical measure time vector is used as the benchmark, the same kind of products’ corresponded degradation measurement vectors are calculated using GA-WSVR models. The Euclid distances of the specific individual and the same kind of products are used to determine degree of membership, so the individual’s degradation path model is built on the basis of degree-of-membership weighted method. Finally, the measurement vectors, Euclid distances, degree of membership and degradation path model are updated with real-time measurement data. The proposed method is applied to fatigue crack growth data, the experimental results validate the validity. References | Related Articles | Metrics

Select

A Simultaneous Strategy Based on Radial Basis Function for Dynamic Optimization WANG Zhi-Qiang, SHAO Zhi-Jiang, WAN Jiao-Na, FANG Xue-Yi 2011, 45 (08):  1221-1225.  Abstract ( 3530 )   Save  A novel simultaneous strategy for solving dynamic optimization problems, which uses the radial basis function interpolation method to discretize the original models, was studied. In this strategy, the inverse multiquadric function is adopted as the basis function, and the indirect approach is used to solve the ordinary equations. All of these concepts were illustrated with two classic test problems successfully.   References | Related Articles | Metrics

Select

Online Detection of Abnormal Data Based on Multilayer Perceptron PAN Yi-Biao-1, YUAN Jing-Qi-1, ZHU Kai-1, CHEN Yu-2, ZHANG Rui-Feng-2 2011, 45 (08):  1226-1229.  Abstract ( 3627 )   Save This work proposed an online detection method of abnormal data based on multilayer perceptron (MLP) and rolling-learning prediction mechanism. In this method, latest historical data with fixed length of data window is used to train an MLP model, and then a one-step-ahead prediction is obtained with the trained MLP model. Secondly, a confidence interval with probability p is calculated with the help of the one-step-ahead prediction and the model residual. New measurement is identified as normal one, if it falls inside the prediction interval; or an abnormal record when it is located out of the prediction interval. Instead of the real measurement, the prediction value is used to update the historical data if abnormal data occurs. Furthermore, through on-line test of real process data collected from a 300 MW coal-fired power generation unit, the effectiveness of the proposed method was verified. References | Related Articles | Metrics

Radiao Electronics, Telecommunication Technology

Select

An Improved Trust Evaluation Model Based on Topology Transformation for Ad Hoc Networks WANG Kun-a, b , GUO Huang-a, b , WU Meng-b, YANG Zhen-b 2011, 45 (08):  1230-1235.  Abstract ( 3500 )   Save  In term of the influence of fake recommendation on the accuracy of recommendation trust in wireless Ad Hoc networks, a trust model based on topology transformation was proposed. Mathematic analyses and simulations results reveal that the proposed trust model has convergence characteristic, the fake recommendation-resisted model performs better in terms of packet successful delivery ratio and mean number of packets dropped. Meanwhile, this enhanced trusted model is more suitable for large scale networks. References | Related Articles | Metrics

Electrotechnology

Select

Low Speed Direct Field Oriented Control on Induction Motor PAN Hai-Peng, GU Li-Fei, GU Min-Ming 2011, 45 (08):  1236-1239.  Abstract ( 4437 )   Save The direct field oriented control was adopted to improve motor’s low speed problem of speed ripple. Based on this method, the combination way of current model flux observer and emf was used to estimate the flux amplitude and thd phase of rotor, and the speed feedback loop formed by current model was used to estimate the rotor speed. As is showed in the simulation results, the method of direct field oriented control of induction motor can improve the dynamic response speed of motor, make motor have better parameters insensitivity, and acquire better static，dynamic performance when motor moves at low speed. The method of flux observer combining current model and emf can make induction motor speed sensorless control system have good static, dynamic performance at high and low speed, which increases the effective speed range of the motor. References | Related Articles | Metrics

Select

Stability and Capacity Analysis of Power Quality Conditioner in Microgrid LI Jun-1, 2 , LIU Xiao-Ning-2, ZHANG Guang-Ming-1 2011, 45 (08):  1240-1245.  Abstract ( 4350 )   Save  According to the power quality problem of microgrid based on new energy distributed generation, this paper introduced a kind of comprehensive power quality conditioner scheme. In the scheme the conditioner series part is used to suppress distributed new energy generation intermittence, and the series side is used to eliminate the harmonic current brought by nonlinear load into the microgrid. On this basis, the paper analyzed the stability of the system and the relationship between  the conditioner capacity and related parameters, and  simulations were performed. The results show that the equivalent parallel resistance is an important factor on the system stability; the voltage volatility is a key factor to affect seriesparallel side conditioner capacity. At the same time, the paper provide the condition of minimum capacity. The results provide guidance for the actual device design and related parameter selection and have  certain theoretical significance and practical value for further study. References | Related Articles | Metrics

Communication and Transportation

Select

Optimal Onoff Control of VMS for Highway Route Guidance SUN Wei-Li-1, LI Ping-1, WANG Hui-1, LIU Hong-2 2011, 45 (08):  1246-1250.  Abstract ( 3416 )   Save The problem of designing route guidance strategies for highway networks via onroute recommendation of variable message signs (VMS) was considered. A generic problem formulation was presented in the format of an optimal onoff control problem whose control action is the switching of VMS. The cell transmission model was then employed to perform network dynamics and a genetic algorithm-based problem solving approach was given. The simulation shows that the strategy with optimal VMS control trajectories can make better use of the highway capability and reduce regional congestion, especially in case the traffic network has unbalanced demand. The total time spent (TTS) of the network can be improved up to 32.39%. References | Related Articles | Metrics