Xiyue Huang

Prediction for chaotic time series based on discrete volterra neural networks

In this paper, based on the Volterra expansion of nonlinear dynamical system functions and the deterministic and nonlinear characterization of chaotic time series, the discrete Volterra neural networks are proposed to make prediction of chaotic time series. The predictive model of chaotic time series is established with the discrete Volterra neural networks and the steps of the learning algorithm with discrete Volterra neural networks are expressed. The predictive model and the learning algorithm are more effective and reliable than the adaptive higher-order nonlinear FIR filter. The Experimental and simulating results show the discrete Volterra neural networks can be successfully used to predict chaotic time series.

Hierarchical fuzzy logic traffic controller for urban signalized intersections

Dynamic signal control is considered an effective measure to deal with urban traffic congestion by increasing intersection capacity and decreasing delays at the same time. A hierarchical fuzzy logic controller for urban signalized intersections is developed in this paper. The controller is designed to be responsive to real-time traffic intensities. Vehicle detectors are placed upstream of the intersection on each approach to measure approach flows and estimate queues. The data is used to decide how much to extend the current green phase. These decisions are made by using a hierarchical fuzzy logic procedure. In the first stage, detected approach traffic flows and the vehicles in the queue are used to estimate relative traffic intensities in the competing approaches. These traffic intensities are then used in the second stage to determine the extension of the current green phase. The performance of this controller is validated in a four-phase signalized intersection by a simulation experiment. The result shows that this control method outperforms the conventional single-stage fuzzy logic controller, especially in the case of heavy traffic flows.

Target Recognition of FLIR Images on Radial Basis Function Neural Network

The study of small target recognition in low SNR (Signal Noise Ratio) is the key problem about processing of forward-looking infrared (FLIR) images information. Eight features of objects based on IR radiation characteristics and wavelet-based are presented. These features are used to a radial basis function (RBF) network as input for learning and classification. The propose recognition algorithm is invariant to the translation, rotation, and scale channel of a shape. Experiments by real infrared images and noisy images are performed, and recognition results show that the method is very effective.

The design and realization of an improved chaotic encryption system

Information becomes more and more important than any time. Now we are living in a data era with development of computer technology. The encryptions were accepted as the most important way to keep the data safe. In this paper, the traditional one-time-key encryption algorithm is given new meaning, and combined with chaotic algorithm become a new encryption. Then the realization of software and hardware of this new encryption proves this method is feasible and effective.

Extension Neural Network Based on Immune Algorithm for Fault Diagnosis

In this paper, the extension neural network (ENN) is proposed.To tune the weights of the ENN for achieving good clustering performance, the immune algorithm(IA) is applied to learning the ENNs weights, which is replaced the BP algorithm. The affinity degree between the antibody and the antigen is measured by extension distance (ED), which is modified to the conjunction function(CF) in Extensions. The learning speed of the proposed ENN is shown to be faster than the traditional neural networks and other fuzzy classification methods. Moreover, the immune learning algorithm has been proved to have high accuracy and less memory consumption. Experimental results from two different examples verify the effectiveness and applicability of the proposed work.

A Bridge-Ship Collision Avoidance System Based on FLIR Image Sequences

In this paper, a forward-looking infrared (FLIR) video surveillance system is presented for collision avoidance of moving ships to bridge piers. An image preprocessing algorithm is proposed to reduce clutter background by multi-scale fractal analysis, in which the blanket method is used for fractal feature computation. Then, the moving ship detection algorithm is developed from image differentials of the fractal feature in the region of surveillance between regularly interval frames. When the moving ships are detected in region of surveillance, the device for safety alert is triggered. Experimental results have shown that the approach is feasible and effective. It has achieved real-time and reliable alert to avoid collisions of moving ships to bridge piers.

Fault Diagnosis Based on Covariance Constraint for Complex Control System

This paper presents the covariance constrain method for fault diagnosis of complex control systems. First of all, the state observation values of fault-spot at the subsystems are extracted in accordance with the operation of the system. Then the average and error of state observation values are defined and the covariance is obtained. Secondly, the weights of various factors are obtained by entropy method. The threshold value is confirmed from the Euclidean norm and the linear weight. When the threshold value is exceeded by the practice data, the alarm will be given and the corrective measures are adopted. Finally, the example shows that the fault covariance constraint control diagnosis method based on the Euclidean norm can be applied to the leakage diagnosis of aerospace products. Comparison the method with the conventional fault diagnosis shows the threshold value mode can be built objectively to realize the real-time fault diagnosis.

Fault Diagnosis and Prediction Based on Hybrid Approach of Wavelet Packet and Extenics

A hybrid wavelet theory and the extenics set theory approach to diagnose and forecast the fault is proposed in this paper. Firstly, the signal is decomposed by the wavelet packet .The feature is extracted from the signal and the repository is established based on the feature of the fault signal. Next ,the real time signal date is took from the sensor and is decomposed by the same wavelet. Secondly, the fault matter element and evaluation matter element is established based on the repository and the real time signal. The conjunction function is connected with the fault matter element and evaluation matter element and the fault of the real time signal is determined by the conjunction function value, then the corresponding measure is took by the probability fault