Journal of Harbin University of Science and Technology 

About: Journal of Harbin University of Science and Technology is an academic journal. The journal publishes majorly in the area(s): Artificial neural network & Wavelet. It has an ISSN identifier of 1007-2683. Over the lifetime, 604 publications have been published receiving 1150 citations.

A Modified Particle Swarm Optimization

Abstract: Particle swarm optimization is a new computational method for tackling optimization functionsHowever,it is easily trapped into the local optimization when solving high-dimension functionsTo overcome this shortcoming,a new particle swarm optimization which improves particle's velocity and position update rule to adjust its movement based on the individual best position is proposed in the paperThe modified algorithm can enhance capability of optimizationFive benchmark functions are tested,and the results indicate that the modified particle swarm optimization is effective to find the global optimal solution

Research of Face Detection Based on Adaboost Algorithm and OpenCV Implementation

Abstract: To solve the problem that the face detector is slow and its rate is low in complicated background,in this paper,we choose a face detector algorithm based on statistics.The principle and training flow of Adaboost face detector is detailedly introduced.The module combine imaging processing software OpenCV and Adaboost face detector algorithm is used to select extended Haar characteristics,then the face detector is realized and the result is given.62 faces is undetected when detector designed for 592 individuals face,and it can reatch about 20 frame/s.Therefore the method of more accuracy and real time.

RHEOTENS Detection and Control System Based on Virtual Instrument Technology

Abstract: In order to test the elongational behaviour of polymer melts and attain the goals of high quality in industry,the paper introduces a system of the measurement of RHEOTENS based on virtual instrument technology.The procedure,method of implementation and hardware structure,software structure are given.The soft-panel of virtual instrument with multi-function is obtained by use of the integrated design idea of virtual instrument and of the environment of visualized graphic programming.The hardware configuration of the system is simple,the software design flexible and the GUI(graphical user interface)friendly.

Study of Fuzzy PID Controller for Industrial Steam Turbine Governing System

Abstract: Since the electric and the heat load of the stream turbine always have uncertain factors,the simple linear or nonlinear differential equations can not sufficiently represent corresponding practical system.And therefore many conventional control methods based on mathematical models are limited.The main purpose of this study attempts to develop a nonlinear fuzzy PID controller adopting fuzzy rule and inference to adjust PID parameters on-line.The proposed fuzzy controller improves dynamic performance of control system greatly and shows excellent robustness against variations of system parameters and external disturbances by comparison with a PID and a fixed fuzzy control.

AHP-ANN Based Credit Risk Assessment for Commercial Banks

Abstract: The goal of this research is promoting the ratio index system and improving the efficiency and quality of credit risk assessment model in the commercial banks Based on the classic credit risk management theory, combining the ANN risk assessment technology with the AHP mathemaiic models, this research sets up the AHP-ANN model for credit risk assessment in commercial banks The result shows, the improved AHP -ANN model promotes the concise index system, criterion measurement and model efficiency in some degree