2003 Vol. 29, No. 9

Display Method:
Preferred growth of nanosized crystal silicon in doped hydrogenated nanocrystalline silicon film
Wei Wensheng, Wang Tianmin, Zhang Chunxi, Li Guohua
2003, 29(9): 753-758.
Abstract:
The preferred growth of nanosized crystal silicon (nc-Si) in doped hydrogenated nanocrystalline silicon (nc-Si:H) films which were prepared using plasma enhanced chemical vapor deposition (PECVD) system by RF(13.56 MHz) and DC bias stimulating was found. Its microstructure was investigated using X-ray diffraction (XRD), Raman spectrum meter, atomic force microscope (AFM) and high resolution transmission electronic microscope (HRTEM). Nc-Si about 10 nm was distributed in the silicon-hydrogen amorphous networks in the films. The peaks of XRD spectra of phosphorus doped nc-Si:H films was at 2θ≈33°, the exponent of crystalline plane was (130). The peaks of XRD spectra of boron doped nc-Si:H films was at 2θ≈47°, the exponent of crystalline plane was (220), respectively. The considerable reasons were investigated. The relation between size of nc-Si, crystalline volume fraction and depositing parameters were studied.
Investigation on methods of kerosene/oxygen ignition using gas dynamic resonance technique
Yu Nanjia, Zhang Guozhou, Ma Bin
2003, 29(9): 759-762.
Abstract:
In order to realize the multiple ignitions of liquid rocket engines reliably, gas dynamic resonance ignition was applied to ignite kerosene/oxygen. Two ignition methods of kerosene/oxygen igniter, direct ignition and indirect ignition, were discussed. For direct ignition method, the differences between the co-axial injection of kerosene/oxygen and the top injection of kerosene were compared. For indirect ignition method, the condition of fuel-rich ignition torch was studied in contrast with the condition of fuel-lean ignition torch. Accordingto the analysis of technical feasibility and difficulty, it can be concluded from experiments that both methods are valid for ignition, while indirect ignition method with fuel-rich torch has more advantages and is prospective to realize in engineering application within a short term.
Research for the kinematic and dynamic models of a flexible glass-wall cleaning robot
Wang Wei, Zong Guanghua
2003, 29(9): 763-766.
Abstract:
A new structure for a smart wall-climbing robot was propossed.The conceptions of Lie Group and differential geometry were used in the robot's controllability analysis. The kinematic and dynamic models of the robot were concluded.The expressions of Lagrange multipliers were acquired to represent the measure of the constraint force.
High performance single reflector compact range
Quan Shaohui, He Guoyu, Xu Yongbin, Jiang Xianzuo
2003, 29(9): 767-769.
Abstract:
A single reflector compact range (CR) was presented. The reflector itself was made up of individual accurately fabricated panels. The precise forming of each panel was based on "nails die, vacuum and honeycomb sandwich" method. Average surface accuracy of these panels was 25 micron rms and surface accuracy of the reflector as a whole was 40 micron rms. Through optimizing electrical geometrical layout, reflector edge design and feed design, high performance was acquired. Practical testing results at 39GHz were presented.
Analysis and application of unsymmetric composite structure
Zhang Xiaodong, Zhang Jikui, Li Zhengneng, Xu Dewei
2003, 29(9): 770-773.
Abstract:
Unsymmetric composite structure is a kind of orthotropic material, and has more designabilities than symmetric composite structure. Those abilities can be used to solve such wing's problems in design as twist diversion, redistribution of load and low efficiency of control. Based upon the results of experiment and calculation of several representative unsymmetric composite laminates, the coupling characteristics of unsymmetric laminate was studied. An airplane wing made of unsymmetric composite structures was designed. And good result was got.
Research on modular combined modeling for parallel mechanism
Zhu Yu, Meng Biao, Zhang Hua, Zhang Yongzhong
2003, 29(9): 774-778.
Abstract:
Based on the combined modeling and a method of describing in character strings of parallel mechanisms, the principium and algorithm of automatically combined modeling of parallel mechanisms were discussed. The modules of automatically combined modeling of parallel mechanisms in PKM-VPDS made the digitized design and modeling of parallel mechanism more efficiently.
Consideration in the structure design of brushless DC motor for flight control
Wang Ziqiang, Zhu Yaozhong, Chen Jiefeng
2003, 29(9): 779-782.
Abstract:
Development of the electromechanical actuation system for use in primary flight control is one of the most technologies in the path to an all-electric airplane. Brushless DC motor is the key in the system. For obtained the high electrical reliability, high mechanical reliability and thermal reliability of the critical issues of this high speed large power density servo motor, the non-magnetic could in special form, dual stator winding, brushless resolver and axial ventilation duct techniques in the structure design was adopted. The result was hoped. Finally, the experimental results of prototype were also offered.
A kind of approach based on signal model for resource configuration in ATE
Zhang Wei, Shen Shituan, Li Yihua
2003, 29(9): 783-788.
Abstract:
Aiming to alleviate the limited test resource situation and provide an automatic configuration process, a signal modeling method based on object-oriented analysis was proposed with a general description language, and according to this consideration, an implementation approach that can acquire resource related information from a given signal was discussed. On the base of the possible relations between resources and signals, the recurrence expressions were presented, and the signal's resource condition and inherit priority were put forward to evaluate the cost and feasibility for the signal's generation and configuration.
Analog circuits fault diagnosis based on serial support vector multi-classifier
Wan Jiuqing, Li Xingshan
2003, 29(9): 789-792.
Abstract:
A new support vector multiclassification methodology was proposed where several binary support vector binary classifiers, each of which equipped with a feature extractor based on kernel principle components analysis (Kernel PCA), were organized in a serial structure. Its training and classification algorithm was described. The BP net classifier, RBF net classifier, traditional support vector multiclassifier and serial support vector multi-classifier (SSVC) were used for analog circuit fault diagnosis. Compared with BP net and RBF net classifiers, support vector approach led to significantly better classification accuracy on test patterns. The SSVC afforded top diagnosis accuracy among these classifiers and outperforms traditional support vector multicalssifier dramatically in training and classification efficiency.
Research and implementation on the intelligent driving behavior model of autonomous virtual human
Qin Shuang, Liu Jinghua, Wen Wenbiao, Zheng Guolei
2003, 29(9): 793-796.
Abstract:
Combining artificial intelligence with virtual reality to create autonomous virtual character is an important aspect of virtual reality research. The intelligent driving behavior of the autonomous virtual human in the virtual traffic environment were concentrated. A hierarchical driver behavior and the perception of the virtual human were descried. Inspired by research on driving behavior, an intelligent driving behavior model based on the production rules for autonomous virtual human was developed. Validated by virtual reality tools, this model was prove to be feasible.
Numerical optimization in a structured/unstructured grid
Tuo Shuangfen, Zhu Ziqiang, Chen Yingchun
2003, 29(9): 797-801.
Abstract:
Much attention had been shifted to the unstructured grid methods. It was due to the relative ease with which these methods could be applied to solve flow problems around complex geometries and the ease with which local refinement of an unstructured grid could be carried out. Numerical optimization and inverse design computation with the viscous flow field were carried out on a structured/unstructured grid. Both mono-objective/bi-objective aerodynamic optimization were included in the numerical optimization. The conventional inverse design, in which the airfoil shape was modified by the required target pressure distribution imposed on an initial airfoil, was realized by present optimization method. Numerical results illustrate that the present method is an efficient tool for airfoil design.
Research on feature extraction of impulse signal using Walsh transform
Li Baoan, Li Xingshan, Liu Xing
2003, 29(9): 802-806.
Abstract:
Compared with Fourier transform, the amplitude of the impulse signal power spectrum of Walsh transform was larger than that of Fourier transform, and the calculation speed of Walsh transform was faster. Wavelet packet analysis could reduce noise of the signal and filter unnecessary signals and characterization of fault signal was put forward. With Walsh transform and wavelet packet analysis, characterization of fault better could be obtained. Bearing vibration signal included fault frequency features when rolling bearing had fault. The Walsh power spectrum of the rolling bearing with fault signal was calculated by using the Walsh transform and discussed for bearing fault diagnosis. Both simulation and testing results indicate that Walsh transform and wavelet packet analysis are effective method in the diagnosis of bearing various faults.
Study on the fatigue properties of the TA15 titanium alloy plate and electron beam welding specimen
Wang Xiangming, Liu Wenting, He Xiaofan
2003, 29(9): 807-811.
Abstract:
Based on the analysis and processing of the fatigue test results of the standard specimens of the TA15 Titanium alloy thin plate and the fatigue test results of the plate at two stress levels, the S-N curve of TA15 plate was established. The fatigue tests at two stress levels were carried out for TA15 electron beam welding plate. The equivalent theoretical stress concentration coefficient of the welded specimens, the mean equivalent theoretical stress concentration coefficient and the reliable equivalent theoretical stress concentration coefficient were determined. Then, the S-N and p-S-N curves for the welded specimens were established. The equivalent Kt method was proved reasonable. Finally, a comparison was made between the fatigue properties of these two kinds of specimens.
Subdivided locating method of star image for star sensor
Wei Xinguo, Zhang Guangjun, Jiang Jie
2003, 29(9): 812-815.
Abstract:
The basic principle of star sensor and major pre-processing methods of star image were introduced. The traditional centroid, threshold centroid, squared centroid and Gauss curve fitting methods were applied into star locating of star image for star sensor. The simulation was carried out sufficiently. The simulation results show that threshold centroid is the most appropriate method for star locating. Simulation in the influences of noise level, filter processing and threshold choosing on the precision of star locating were made. The star locating of the actual star image was done based on threshold centroid, and the satisfying result was obtained.
Experimental investigation of high temperature fatigue properties of GH2132
Niu Huipeng, Zhang Zheng, Liu Shaolun
2003, 29(9): 816-820.
Abstract:
High cycle fatigue and low cycle fatigue properties and fracture behavior of GH2132 were investigated under different temperature, stress intensity factor and stress. The effect of each parameter was analyzed to determine the interaction on fatigue life. It was found that the parameter, which affected the life mostly, was different at different experimental conditions. Combining results by scanning electronic microscopy, the reason of the interaction was revealed.
Research on the modeling and algorithm of multi-objective crew rostering
Li Qing, Zhang Jun, Zhang Xuejun
2003, 29(9): 821-824.
Abstract:
To improve the solution of the rostering problem, a multi-objective optimization model was proposed. The adaptive genetic algorithm based on entropy was improved and was used to solve the rostering problem to attain the best solution. In the improved method, inferior individuals were adopted with some probability as simulated annealing. Individuals of next generation were selected by using set partitioning method. The correctness and advancement of this model and algorithm were tested by solving aircrew rostering problem of Yunnan Airline.
Real-time algorithm for the detection and classification of the exceptions in the home security surveillance
Zhou Bing, Du Runqiu, Li Bo
2003, 29(9): 825-828.
Abstract:
Fast and robust detection of moving objects was difficult for a security surveillance system based PC. By combing the illumination and the chroma in the YUV color space, the problem was solved that a big breaking-in object causes the background pixels to change its illumination. A fast method for analyzing and classifying the moving objects was introduced, based on some detection such as light switching, moving distance and skins. In order to eliminate swaying objects in the background and to reject false alarms, a new method of background maintenance was proposed.
Development platform for embedded graphics user interface
Xing Weiwei, Fan Shangchun
2003, 29(9): 829-833.
Abstract:
According to the requirements and characteristics of products as intelligent instruments based on MCU, DEG, a development platform for low-side embedded GUI systems, had been developed. DEG was a collection of development tools and graphics library. The graphics library, which was organized as an object-oriented application framework, was key for code reusing and platformization of embedded systems development. The framework offered the macroscopical framework and structural elements that was necessary to build a simple multi-window GUI systembased on MCU. Its macroscopical framework was organizied as hierarchy system, and run under event driving mechanism. According to the characteristics of MCU, organization mode of the class library was a flat interface-inherit mode, insteadof the complex class-inherit mode. DEG had been successfully applied into the development of the GUI sub-system of an intelligent instrument product.
Implementation of elliptic curve digital signature arithmetic based on smart card
Zhang Jinghe, Zhang Qishan, Wu Jinpei
2003, 29(9): 834-837.
Abstract:
Elliptic curve digital signature arithmetic can achieve high security using a shorten key, so it is very suitable to realize this arithmetic on smart card. The characteristic of hardware accelerator was introduced. In order to increase the executive speed of elliptic curve digital signature arithimeitc, projective coordinates was adopted in stead of affine coordinates. The executive time was provided .The result indicates that the arithmetic can work availably, comparing with RSA and DSA ,the executive time of elliptic curve digital signature arithmetic is reduced greatly.
Multiuser detection based on hidden Markov models
Liu Feng, Zhang Taiyi, Hu Jinyan
2003, 29(9): 838-842.
Abstract:
For investigation on the slow convergence speed of channel estimation and the unreason of state estimation in the probabilistic algorithms, a new multiuser detection was presented. Based on the hidden Markov models, it adopted the fixed-lag to avoid divergence in the estimation channel, and used the sawtooth-lag scheme in calculating the probability of the backward sequence to decrease computational complexity. With the correlation between the previous and the following sequence and the steepest descent algorithm, the new algorithm utilized DLMS to estimate the mutlichannel of the multiuser. Considering seeking for the following sequence of maximal transition probability corresponding to previous sequence, a maximum a posteriori (MAP) multiuser detector had been constructed. The computer simulation illustrates that the new algorithm increase the speed of channel estimation, global convergence and system stability.
Research on the method of simulating road roughness numerically
Liu Xi, ong, Deng Zhidang
2003, 29(9): 843-846.
Abstract:
To obtain exact road roughness, a new method processing directly the given power spectral density was introduced. Considering the effect of the vehicle natural vibration frequency and velocity completely, the simulation and analysis were done on the C grade of road using this method. The results indicate that the power spectral density of the road roughness obtained is the same as the given.