Application of Improved Differential Evolution Algorithm in solving Container-packing problems
Application of Self-Adaptive Wavelet Neural Networks in Ultrasonic Detecting of Drainpipe
Download 1.34 Mb.
|
- Navigate this page:
- Implementation on Google Maps-style WebGIS based on ArcGIS
- Identity Authentication System Based on Improved PR-RP Model
- Digital modulation identification by distance and fuzzy evaluation
- Adaptive fuzzy clustering based on genetic algorithm
- Application of Hydraulic System in Condition Measurement and Fault Diagnose Based on of Information Fusion
- Research of Enterprise Application Integration Based-on ESB
- Research on Dynamic History Curve Based on Embedded Web Server
- A Data-driven Multiple-model Modeling Algorithm
- Using Dual-gene Variation Hybrid Evolution Strategy to Construct Fire Emergency Decision-making Model
- A new Method of Seismic Wavefield Separation and Denoising
- Semi-Supervised Nonlinear Dimensionality Reduction with Pairwise Constraints
- Hybrid Particle Filter for Vehicle MEMS-INS
- Model Analysis of the Time-Sharing Mode
- An algorithm of target tracking based on adaptive LFM waveform design
- Research on application of Distributed Data Mining in Anti-Money Laundering Monitoring System
- Fault Trend Analysis of Rotating Machine Based on Degree of Cyclostation
- Recognize the Low-efficiency Water Passage in Multilayer Non-uniform Reservoir with Fuzzy Evaluation
- A method for the shortest distance routing in a chart based on testing a spatial relation of a route segment and an obstacle area
Application of Self-Adaptive Wavelet Neural Networks in Ultrasonic Detecting of Drainpipe Xi-Peng Yin Department of Electronic Engineering Northwestern Polytechnical University, NPU Xi’an, Shaanxi, P.R.China yinxipeng@nwpu.edu.cn
Department of Electronic Engineering Northwestern Polytechnical University, NPU Xi’an, Shaanxi, P.R.China Fan_yangyu@nwpu.edu.cn Abstract—Drainpipe ultrasonic non-destructive testing is liable to be interfered with the external environment. So it is important to remove the noise signal effectively in drainpipe ultrasonic non-destructive testing. The testing system is constructed by self-adaptive wavelet neural networks which is using the wavelet and neural network algorithm. Better fitting signal is achieved by choosing Orthogonal Daubechies wavelet neuron and optimizing the scale parameter. The simulation results showed less distortion and better noise cancellation. Keywords- ultrasonic; self-adaptive; wavelet analysis; neural networks
TAN Qingquan, QIAO Yongjun, WANG Zhanying Earthquake Administration of Beijing Municipality No.28, Suzhou Street, Haidian District Beijing, China 100080 irsatanqq@126.com LIU Qun* School of Electrical & Electronic Engineering, Shandong University of Technology No.12, Zhangzhou Road, Zhangdian District Zibo, China 255049 dkliuqun@sdut.edu.cn
Keywords-WebGIS; ArcGIS; Google Maps Identity Authentication System Based on Improved PR-RP Model Zhang zhi-yuan1,2, Xiao chuang-bai1, Zhao de-qun1,3 1. College of Computer Science and Technology, Beijing University of Technology, Beijing 100124 China 2. Department of Computer and Information Engineering, Changsha Telecommunications and Technology Vocational College, Changsha Hunan 410015, China 3. Department of Computer Engineering, XinJiang polytechnich college, Urumqi 830091, China cbxiao@bjut.edu.cn; zzy@csctc.net; Sun hong1, Xie Kai1, Tian guo-zhong1,2*
Urumqi 830091, China 2. College of Computer Science and Technology, Beijing University of Technology, Beijing 100124 China tiangz@emails.bjut.edu.cn Abstract
Key Words: Keystroke Rhythm; PR-RP model; Characteristics Template; Identity Authentication Digital modulation identification by distance and fuzzy evaluation Wang Lanxun Wu Honglin School of Electronic & Information Engineering Hebei University Baoding China Email: wanglanxun@yahoo.com.cn holypla@126.com Abstract—In order to overcome the problem of erroneous identification for using fixed threshold of the characteristic parameters of signal to identify signals, this article extracted several parameters and combined with the feature distance and fuzzy evaluation to identify the signal. It is divided into two levels, first of all, comparing the distance to identify the inter-class of signal, and then use the characteristic parameters and fuzzy evaluation to identify the type of signal. This article analyses the effectiveness of the method very closely, and verify the validity in identifying MPSK and MFSK signal through computer simulation. Keywords: modulation recognition; Feature distance; fuzzy evaluation; MPSK and MFSK
Zhu Lianjiang College of Information Science and Engineering University of Jinan Jinan, China ise_zhulj@ujn.edu.cn Qu Shouning Information Network Center University of Jinan Jinan, China qsn@ujn.edu.cn
College of Information Science and Engineering University of Jinan Jinan, China eo_dut@ujn.edu.cn ABSTRACT: Traditional Fuzzy c-means (FCM) algorithm is commonly used in unsupervised learning. However, there are some limitations. Cluster number should be determined and the cluster center should be initialized before classification. A new algorithm is proposed in the paper. The best cluster number is obtained by analyzing cluster validity function and the cluster center is initialized by HCM. The data set is classified with Fuzzy c-means algorithm based on Genetic algorithm. The experimental results indicate the effectiveness and adaptability of the new algorithm. KEYWORDS: Cluster Analysis; Cluster validity; Fuzzy C-means; Genetic algorithm
LIANG Guihang school of traffic ludong university Yantai, China liangguihang@163.com YU Jingnuo school of traffic ludong university Yantai, China yujingnuo@126.com MU Chunyan school of traffic ludong university Yantai, China muchunyan@163.com FENG Baofu school of traffic ludong university Yantai, China fbf@fzu.edu.cn
Keywords-information fusion; fuzzy neural network; condition measurement; fault diagnosis Research of Enterprise Application Integration Based-on ESB Jieming Wu Department of Computer Science North China University of Technology Beijing, China wujieming@263.net Xiaoli Tao Department of Computer Science North China University of Technology Beijing, China taoxiaoli1012@126.com
Keywords- EAI;ESB;SOA Research on Dynamic History Curve Based on Embedded Web Server Xinping Liu Computer and communication engineering China University of Petroleum Dongying, China e-mail: liuxinp@upc.edu.cn Qingzhi Wang Computer and communication engineering China University of Petroleum Dongying, China e-mail: wangqingzhi0617@126.com Abstract—Clients can exchange data with embedded device where Web server is transplanted. In order to observe the state of the device and display it in the form of dynamic history curve, this paper has adopted two methods to get the real-time data of the embedded device and display the curve. One uses JavaScript cooperating with VML (Vector Markup Language), the other uses Applet. The two methods both have good displaying effect. The first method is complicated. It demands that clients install IE browser. The code of the second method is simple but needs the installation of JRE on the client side. These two methods have a quite bright prospect and can be applied in various fields. Keywords- embedded Web server; dynamic history curve; Applet; VML; CGI
Jikun Ye,Humin Lei,Fei Wang ,Jiong Li, Lei Shao Dept. of Control Science and Engineer Missile Inistitude of Air Force Engineer University Sanyuan,Shanxi e-mail: jikunbo@sina.com Abstract—Based on the input-output data , a multiple-model modeling method is suggested for the complex nonlinear system. Firstly, fuzzy partition is employed to on-line clustering for the input-output data; and then, the least-squares (LS) algorithm is employed to construct the local model for each clustering, and the parameter of each local model is updated by the new data. The proposed algorithm takes advantage of the TSK model, combines the fuzzy partition and multiple-model modeling, and updates on-line the number of the local model and the parameter of each by the input-output data, so as to realize the on-line modeling for the complex nonlinear system. Simulation result shows the effectiveness of the proposed method. Keywords- nonlinear system; multiple models; on-line modeling; clustering; fuzzy partition; Using Dual-gene Variation Hybrid Evolution Strategy to Construct Fire Emergency Decision-making Model Hongwei Xie College of Computer and Software, Taiyuan University of Technology， Taiyuan, China(030024 ) xiehongwei@tyut.edu.cn Li Zhang College of Computer and Software, Taiyuan University of Technology， Taiyuan, China(030024 ) zhangl_101@tom.com Abstract—The time parameter about emergency decision-making problems is the primary optimizational object. Evolutionary strategy (ES) can solve the problems of continuous parameter optimization, and is an effective way of solving complex optimizational problems. Therefore, in the paper dual-gene variation hybrid evolution strategy algorithm is used to construct the fire emergency decision-making model, in order to solve parameter optimizational problems. At the same time, the convergence speed of dual-gene variation hybrid evolution strategy algorithm is verified. Keywords-dual-gene variation hybrid evolution strategy algorithm; fire emergency decision-making model
School of Petroleum Resources Xi'an Shiyou University Xi’an,China S_hongyan@sohu.com Qing-chun Li College of Geology Engineering and Geometics Chang’An University Xi’an, China dcliqc@chd.edu.cn Abstract—Singular Value Decomposition（SVD）Filtering is one seismic data processing technique which uses the lateral coherence difference of the seismic wave to achieve wavefield separation and denoising. As the differences of propagation characteristics, apparent velocity and coherence in seismic signals, transformation of the valid signals may be transformed into those with better coherence by some mathematic transformations. In this paper, based on apparent velocities, throughing the processing method of two times Normal Moveout(NMO), we make the P-P wave and the P-S wave moveout respectively that enable them to achieve the best coherence on the deflection. The goal is that transforms the desired signal to one kind better processing territory in the horizontal upward coherence, then distinguishes two times SVD respectively and reconstruct signal with extracting the singular value of target signals. we might accomplish P-P wave and P-S wave separation as well as noise attenuation afterwards. Keywords- Seismic data processing; Normal Moveout (NMO); Singular Value Decomposition（SVD）; P-P wave; P-S wave; Wave field separation and denoising racking Control Research of High-Order Flexible Structures on The H-infinity Control Method YOU Wei-qian, CHEN Huai-hai, HE Xu-dong MOE Key Lab of Structure Mechanics and Control for Aircraft; Institute of Vibration Engineering Research, Nanjing University of Aeronautics and Astronautics, Nanjing, China. Email: youweiqian@gmail.com Abstract: The H-infinity mixed sensitivity tracking control method is applied to the vibration control of high-order flexible structures and a method for the choice of the mixed sensitivity weighting matrix is put forward. A MIMO (multi-input multi-output) control model of a vibration system is established which induced model uncertainties via the H-infinity control theory and the mixed sensitivity design method is used to design the robust H-infinity controller. An augment plant for the vibration system is presented by introducing the weighting function matrices. A MIMO control model with two inputs and two outputs is set up by a cantilever beam for simulation. The results show that the H-infinity controller designed for the high-order flexible structure can satisfy the requirements of the internal stability of the control system and the tracking specifications. Keywords：H-infinity Control; Vibration control;Tracking control; mixed sensitivity; MIMO system Semi-Supervised Nonlinear Dimensionality Reduction with Pairwise Constraints Min Chen 1 1 Department of Computer Science and Technology Hunan Institue of Technology Hunan, China E-mail: milechen@163.com Zhao Zhang 2 2 School of Computer Science and Technology Nanjing Forestry University Nanjing, China E-mail:zzhang618@gmail.com Abstract— The problem of semi-supervised dimensionality reduction with kernels called KS2DR is considered for semi- supervised learning. In this setting, domain knowledge in the form of pair constraints is adopted to specify whether pairs of instances belong to the same class or not. KS2DR can project the samples data onto a set of ‘useful’ features and preserve the structure of unlabeled samples data as well as both similar and dissimilar constraints defined in the feature space, under which the samples with different class labels are easier to be effectively partitioned from each other. We demonstrate the practical usefulness and high scalability of KS2DR algorithms in data visualization and classification tasks through extensive simulation studies. Experimental results show the proposed methods can almost always achieve the highest accuracy when the dimension is reduced. And KS2DR methods outperform some established dimensionality reduction methods no matter how many numbers of constraints, dimensions are used. Key Words— Semi-supervised learning; Kernel feature space; Dimensionality reduction; (Dis-) similar constraints
Lijie Yu
College of Automation, Beijing Union University Beijing 100101, China zdhtlijie@buu.edu.cn Zongyu Gao and Desheng Li MEMS Technology Research Center Beijing University of Technology Beijing 100124, China gzy19750510@163.com Abstract---This paper proposes hybrid unscented particle filter for integrated navigation system of vehicle which have characteristics of long-time moving and frequent changes in dynamic performance and different non-linear and non- Gauss characteristics. The algorithm combined with zero velocity updates method (ZUPT) and effectively reduced the error of different vehicle’s dynamic characteristics based on the integrated navigation system of vehicle. The test result show the method effectively to improved positioning accuracy and the error accumulation of vehicle. Index Terms--- hybrid unscented particle filter (HUPF)；integrated navigation system (INS)；ZUPT；Non-linear；GPS output
Liu Guo-cai Microgravity department Center for Space Science & Applied Research Chinese Academy of Sciences No.1 Nanertiao street Zhong guan cun , Hai dian district Beijing China , 100190 086-010-82646264 E-mail: liu_guocai@cssar.ac.cn Liu Chao-ying College of Electrical Engineering & Informational Science Hebei University of Science & Technology No.70 Yuhua East Road Shijiazhuang, Hebei Province, China, 050018 086-0311-85383057 E-mail:liucy@hebust.edu.cn You Yu-yang College of Computer Science and Technology Harbin Engineering University No.145 Nan tong street Nan gang district Harbin Hei long jiang China , 150001 086-0451-88780471 E-mail: arthurwy@163.com Abstract—Transfer functions of zones and arbitrary on the time-sharing working mode of the multi-zone space crystal growth furnace made in china were constructed. Taking two and three zones with equal time-sharing mode for example, the static characteristics while and transient properties while were offered and Comparisons with zero-order holder were made. At last, conclusions that the time-sharing mode with rich properties and the essence of the time-sharing mode is high dimensional surface on , and . Contributions of theoretical references to complete the control system model of space crystal furnace are offered by studying the properties of time-sharing mode. Keywords-time-sharing mode;property analysis; high dimensional surface An algorithm of target tracking based on adaptive LFM waveform design Zhang Zhen-kai, Zhou Jian-jiang and Wang Fei College of Information Science and Technology NanJing University of Aeronautics and Astronautics Nanjing, China zhangzhenkai@126.com, zjjee@nuaa.edu.cn, wangxiaoxian@nuaa.edu.cn Zhang Zhen-kai College of Electronic Information Jiangsu University of Science and Technology Zhenjiang, China zhangzhenkai@126.com Abstract—An algorithm of target tracking based on adaptive waveform design was proposed with the help of Doppler estimation theory. According to the analysis of the ambiguity function for liner frequency modulation signal, a model between ambiguity function of the signal and Cramer-Rao bound of time and frequency delay was built. This paper realized the target tracking by adjusting the frequency width and pulse width using Kalman filtering method. With the same signal energy and noise environment, the simulation results show that the proposed adaptive tracking algorithm has a better tracking precision than the method which makes use of the constant parameter. Key Words-Adaptivity; Waveform design; Target tracking; Kalman Filtering; Ambiguity function
Cheng-wei Zhang School of Management, Dalian University of Technology Dalian, China, 16000 E-mail: zcwdlut@163.com Yu-bo Wang School of Software, Dalian University of Technology Dalian, China, 16000 E-mail: wangyubo0903@live.com Abstract—From the reality, this article presents anti-money laundering real-time system architecture on the base of analyzing custom transaction. At the same time, data process, custom transaction and bank node are discussed. Key Words-Distributed Data; Anti-Money System; Real Time Monitoring; Data Mining; BI
Xu xiaoli Beijing Information Science and Technology University Beijing, China xuxiaoli@bistu.edu.cn Zuo Yunbo Beijing Institute of Technology Beijing, China 2001zyb@163.com Abstract—Most parts of rotating machine rotate cyclically when the machine fails for some reason, and the vibration signal of rotating machine often has the cyclostation character. Aiming at this feature, Degree of Cyclostation(DCS) was presented to analyze the fault trend of rotating machine in this paper. DCS was constructed by the comparison between third-order non-stationary cyclic moments and stationary cyclic moments. It evaluated the cyclic frequency with the binary decision method and it could reflect the non-stationary character of vibration signal and suppress the noise and interference of the signal. In the experiments of analyzing the fault trend of rotating machine, DCS could get better trend result than the traditional vibration amplitude and vibration intensity. Keywords- rotating machine; Degree of Cyclostation; fault trend; vibration intensity
Liu Li
Key Laboratory of Enhanced Oil and Gas Recovery of Ministry of Education of China, Daqing Petroleum Institute, Daqing, China e-mail: liuliduoduo@163.com Song Kaoping, Pi Yanfu, Yang Jing Key Laboratory of Enhanced Oil and Gas Recovery of Ministry of Education of China, Daqing Petroleum Institute, Daqing, China e-mail: skp2001@sina.com Abstract—For multilayer sandstone reservoir , high permeability water passage would be formed between oil well and water well at the later stage of water drive. This caused the injected water circulation with no or low efficiency and thus decreased the production of oil wells. In order to improve the effect of water drive, measures of profile control and/or water shutoff must be taken. The key problem is to recognize the wells and layers having ineffective or low-efficiency water passage. In this paper, the identifying indexes have been determined and corresponding mathematical model has been constructed with fuzzy theory. Keywords—low-efficiency water passage; water drive; mathematical model; multilayer sandstone reservoir; fuzzy theory
Zhang Li 1 1. Department of Navigation, Dalian Maritime University, Dalian, Liaoning 116020, PR China zhldlny@126.com Zhang Lihua2 Wang Zhu 2 Jia Shuaidong 2
Dalian Naval Academy, Dalian, Liaoning 116018, PR China Abstract—An optimal planned route is a foundation of safe and economical navigation. This paper proposes a method for the shortest distance routing in a chart based on testing a spatial relation of a linear route segment and an obstacle area. Firstly, safety contour traces are traced and obstacle areas are acquired automatically. Then possible optimum paths circumambulating an obstacle are searched automatically by comparing the spatial relation of the linear route segment and the nearest obstacle area. Finally a route binary tree is constructed and the shortest route is obtained. Experimental results demonstrate that the proposed method has more advantageous over traditional methods in efficiency, accuracy and reliability. Keywords- the shortest distance routing; spatial relations; route segment; obstacle area Directory: meeting -> webmedia -> jingpin meeting -> Iea shcp/pvps working Group on pv/t solar Systems meeting -> Office of the collector: jagatsinghpur (st & sc development Section) meeting -> World meteorological organization meeting -> WG/hwsor action item summary (Current as of 28 Feb 2013) meeting -> Review of the past hurricane season meeting -> Review of the past hurricane season reports of hurricanes, tropical storms, tropical disturbances and related flooding during meeting -> Town of seabrook island meeting -> World meteorological organization jingpin -> Modis-based Inner Mongolia Grassland Snow-Cover Mapping Download 1.34 Mb. Share with your friends:
|