Young-Hyun Lee

Development of an Automatic Stencil Inspection System Using Modified Hough Transform and Fuzzy Logic

In this paper, the authors present a system to inspect metal stencil that is used to print solder paste on pads of surface-mounted device on printed circuit board. The developed inspection system is composed of a moderately precise X-Y robot and a vision system. To correct a position error caused by the X-Y robot, the authors define position error vector and apply modified Hough transform to determine the dominant position error vector. Using this extracted dominant position error vector, the reference image is modified. This transformed reference image is compared with the camera image. Fuzzy logic is utilized to judge the correctness of the holes on the stencil. The input variables are the ratio of the overlapped area of two holes and the distance between the centroid of them. The output variable is the grade of the identity of the hole. These methods are verified by a simulation and applied to the inspection system

Development of Real-time Diagnosis Method for PEMFC Stack via Intermodulation Method

Abstract : During PEMFC(Proton Exchange Membrane Fuel Cell) operation monitoring and diagnosis are important issues for reliability and durability. Stack defect can be followed by a critical cell voltage drop in the stack. One method for monitoring the cell voltage is CVM(Cell Voltage Monitoring), where all cells in the stack are electrically connected to a voltage measuring system and monitored these voltages. The other methods are based on the EIS(Electrochemical Impedance Spectroscopy) and on nonlinear frequency response. In this paper, intermodulation(IM) method for diagnosis PEMFC stack is introduced. To detect one or more critical PEMFC cell voltage PEMFC stack is excited by two or more test sinusoid current, and the frequency response of the stack voltage is analyzed. If one or more critical cell voltage exists, higher harmonics on the voltage frequency spectrum will appear. For the proposed IM method, stack simulation and experiments are conducted. Key words : PEMFC stack(양성자교환막 스택), Stack diagnosis(스택 진단), Critical fuel cell voltage(임계 연료전지 전압), Intermodulation(인터모듈레이션)

Development of Intelligent Diagnosis Method Using Fuzzy Logic and Artificial Neural Network for PEMFC Stack

For Fuel Cell Electric Vehicles(FCEV) to attain a competitive advantage today, durability and an acceptable automotive system cost are requisite. The durability of the Proton Exchange Membrane Fuel Cell(PEMFC) stack used in the FCEV is ensured particularly in real-time, on-line diagnosis and optimal control. To address price competitiveness, it is necessary for the main part to minimize the cost of the platinum catalyst that is usually not unavoidable. In this paper, we developed intelligent diagnosis methods by using Fuzzy Logic algorithms and Artificial Neural Networks to improve the existing, real-time diagnosis methods. We also deployed an intelligent diagnosis method on a real FCEV stack to verify the algorithm. To analyze the dynamic characteristics, the HWFET(Highway Fuel Economy Test) driving cycle was used.

Position Estimation of Wheeled Mobile Robot in a Corridor Using Neural Network

This paper describes position estimation algorithm using neural network for the navigation of the vision-based Wheeled Mobile Robot (WMR) in a corridor with taking ceiling lamps as landmark. From images of a corridor the lamp`s line on the ceiling in corridor has a specific slope to the lateral position of the WMR. The vanishing point produced by the lamp`s line also has a specific position to the orientation of WMR. The ceiling lamps has a limited size and shape like a circle in image. Simple image processing algorithms are used to extract lamps from the corridor image. Then the lamp`s line and vanishing point`s position are defined and calculated at known position of WMR in a corridor To estimate the lateral position and orientation of WMR from an image, the relationship between the position of WMR and the features of ceiling lamps have to be defined. Data set between position of WMR and features of lamps are configured. Neural network are composed and teamed with data set. Back propagation algorithm(BPN) is used for learning. And it is applied in navigation of WMR in a corridor.