Time filter

Source Type

Gan X.-S.,XiJing College | Gao H.-L.,Unit 93132 of PLA
Advanced Materials Research | Year: 2014

To improve the learning capability of Radial Basis Function (RBF) neural network, a RBF neural network algorithm based on Extended Kalman Filter (EKF) is proposed. First the basic idea of EKF algorithm and RBF neural network are introduced, and then EKF is used to optimize the parameters combination of RBF neural network to obtain the better model. The experiment proves its feasibility. © (2014) Trans Tech Publications, Switzerland.

Gan X.-S.,Xijing College | Duanmu J.-S.,PLA Air Force Aviation University | Cong W.,PLA Air Force Aviation University
Kongzhi yu Juece/Control and Decision | Year: 2011

For the shortcoming of traditional wavelet neural network(WNN), a WNN algorithm based on modified unscented Kalman filter(UKF) is proposed. The algorithm uses an UKF based on Sigma point of simplex spherical distribution(SSUKF) to train the parameters of WNN, which can improve the learning ability and training quality of WNN. The experiment results on aerodynamic modeling of flight vehicle show that, compared with BP and extended Kalman filter(EKF), the WNN trained by SSUKF algorithm has a better ability with features of convergence, precision and calculation, and is also a good method for aerodynamic modeling of flight vehicle.

Gan X.-S.,XiJing College | Duanmu J.-S.,PLA Air Force Aviation University | Wang J.-F.,PLA Air Force Aviation University | Cong W.,PLA Air Force Aviation University
Knowledge-Based Systems | Year: 2013

To improve the ability of detecting anomaly intrusions, a combined algorithm is proposed based on Partial Least Square (PLS) feature extraction and Core Vector Machine (CVM) algorithms. Principal elements are firstly extracted from the data set using the feature extraction of PLS algorithm to construct the feature set, and then the anomaly intrusion detection model for the feature set is established by virtue of the speediness superiority of CVM algorithm in processing large-scale sample data. Finally, anomaly intrusion actions are checked and judged using this model. Experiments based on KDD99 data set verify the feasibility and validity of the combined algorithm. © 2012 Elsevier B.V. All rights reserved.

Gan X.-S.,XiJing College | Tang X.-Q.,Urumchi | Gao J.-G.,PLA Air Force Aviation University
Applied Mechanics and Materials | Year: 2013

Based on the study of adaptive wavelet neural networks, a hierarchy genetic algorithm is proposed to training network. Compared with standard genetic algorithm, the method can not only optimize network parameters such as scale factor, transform factor parameter and connection weights, but also solve structure problem of adaptive wavelet neural network. The result of simulation indicates that the algorithm can efficiently determinate the parameter and structure of adaptive wavelet neural network, and has better higher training efficiency and forecasting precision. © (2013) Trans Tech Publications, Switzerland.

Zhao H.-C.,XiJing College
Applied Mechanics and Materials | Year: 2013

To improve the performance of Wavelet Neural Network (WNN), a hybrid WNN learning algorithm, which is combination of Genetic Algorithm (GA) and Chaos Optimization Algorithm (COA) in a mutual complementarity manner, is proposed. In the algorithm, GA is first used to roughly search the optimal parameters of WNN as a whole, and then COA is adopted to perform the refined search on the basis of the result obtained by GA, which can make remarkable progress in modeling accuracy, learning speed, and overcoming local convergence or precocity. Simulation show its effectiveness. © (2013) Trans Tech Publications, Switzerland.

Gan X.S.,XiJing College | Li H.-P.,Air Force Xi an Flight Academy
Advanced Materials Research | Year: 2014

To assess the safety situation in air traffic control effectively, a comprehensive assessment method based on RBF neural network is proposed. At first, the safety assessment principle based on RBF neural network is introduced, and then the safety assessment index system for air traffic control is established from four aspects of human, machine, environment and management. Simulation example of ATC safety assessment gives us a satisfactory result for RBF neural network. © (2014) Trans Tech Publications, Switzerland.

Bin O.,Xijing College
Advance Journal of Food Science and Technology | Year: 2013

In view of the enterprise ecological apple manor a variety of problems of existence, put forward to the enterprise management transformation, achieve enterprise, collective, individual integrated operation management and the use of regression mathematical model on apple professional cooperation organization analysis. Through the example, Apple professional economic cooperation organization innovation model of the input output ratio than the rural economic cooperation organization is much higher, the conclusion and the utility of the marginal analysis theory is contrary to. Conclusion the apple professional economic cooperation organization of research and development has a certain reference value. © Maxwell Scientific Organization, 2013.

In order to more effectively solve some difficult problems in gearbox failure diagnosis, a gearbox fault diagnosis method based on Relevance Vector Machine (RVM) is proposed. RVM is developed in Bayesian framework. It does not need to estimate the regularization parameter with less relevance vectors, and its kernel function does not need to satisfy Mercer condition. Simulation results show that: compared with the traditional BP neural network, RVM has the faster modeling speed, more accurate diagnosis, and is worthy of promotion and application in fault diagnosis of the gearbox. © (2013) Trans Tech Publications, Switzerland.

Wang Z.,Xijing College
Kongzhi Lilun Yu Yingyong/Control Theory and Applications | Year: 2011

The passivity control problem for a nonlinear electromechanical transducer chaotic system with relative 2nd order is studied. It is assumed that the system has a standard chain structure. By using the Backstepping method and the equivalence relation between the passivity and the stability, feedback stabilizing controller of the system is designed and has been proved. The simulation results demonstrate the effectiveness of the proposed method.

Bin O.,Xijing College
Advance Journal of Food Science and Technology | Year: 2014

The Markov chain empirical prediction results of economic models based on agricultural enterprises, so as to avoid investment risks. Venture capital investment as a new investment tool, its earnings forecast method is better than the traditional finance investment prediction, which is more complex, more professional. This study constructs a risk investment based on Markov chain prediction of economic model, the fitting matrix method to estimate one-step transfer probability matrix. Finally taking Shaanxi Huasheng Group's actual earnings data as an example, a calculating example is given, which shows the effectiveness of the method. This study presents the economic model for the development of the venture capital industry has certain positive role. © Maxwell Scientific Organization, 2014.

Loading XiJing College collaborators
Loading XiJing College collaborators