http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
자동 8자 위빙 장치를 장착한 알곤 용접토치 및 제어장치의 개발
사의환(Yui-hwan Sa),정성용(Sung-Yong Jung),차형우(Hyeong-Woo Cha) 대한전자공학회 2016 대한전자공학회 학술대회 Vol.2016 No.6
An argon welding torch mounted automatic 8-style weaving device and its the control equipment was development. The welding torch consists of automatic wire supply, 8-style weaving device, conventional torch and control switches. The control system using Arduino can control initial setting up, wire supply, 8-style weaving, and start/stop welding. The fixtures of 8-style weaving device, wire supply, and its control system were also designed and developed.
제3-고조파를 이용한 저항과 용량형 센서 인터페이스 회로 설계
사의환(Yui-Hwan Sa),손표훈(Pyo-Hoon Son),김기홍(Ki-Hong Kim),김희석(Hi-Seok Kim),차형우(Hyeong-Woo Cha) 대한전자공학회 2016 대한전자공학회 학술대회 Vol.2016 No.11
A novel resistive and capacitive interface circuit using third harmonic frequency was designed. The interface consists of fixed and variable pulse-wave oscillators, two band-pass filters with third harmonic center frequency of the two oscillators, a multiplier, and a low-pass filter(LPF). The band-pass filter outs three times capacitive(resistive) -difference frequency of the fixed and variable capacitor(resistor). The difference frequency of the multiplier was filtered by LPF and than final frequency of the LPF was proportioned to the three time of capacitive (resistive)-difference. The experimentation results show that the proposed interface has resolution of 1.59Hz/fF and nonlinearity error of 1.86%.
Cogging Torque Reduction Design for CVVT Using Response Surface Methodology
Jae-Yui Kim(김재의),Dong-min Kim(김동민),Soo-Hwan Park(박수환),Jung-Pyo Hong(홍정표) 대한전기학회 2016 전기학회논문지 Vol.65 No.12
This paper deals with the design process for an outer-rotor-type surface-mounted permanent magnet synchronous motor (SPMSM) used in continuous variable valve timing (CVVT) systems in automobiles with internal combustion engines. When the same size, outer-rotor-type SPMSMs generate larger torque and more stable than inner-rotor-type SPMSMs. For the initial design, space harmonic analysis (SHA) is used. In order to minimize the cogging torque, an optimization was conducted using Response Surface Methodology (RSM). At the end of the paper, Finite Element Analysis (FEA) is performed to verify the performance of the optimum model.
A Fabrication and Testing of New RC CMOS Oscillator Insensitive Supply Voltage Variation
Jin-su Kim,Yui-hwan Sa,Hi-seok Kim,Hyeong-woo Cha 대한전자공학회 2016 IEIE Transactions on Smart Processing & Computing Vol.5 No.2
A controller area network (CAN) receiver measures differential voltage on a bus to determine the bus level. Since 3.3V transceivers generate the same differential voltage as 5V transceivers (usually ≥1.5V), all transceivers on the bus (regardless of supply voltage) can decipher the message. In fact, the other transceivers cannot even determine or show that there is anything different about the differential voltage levels. A new CMOS RC oscillator insensitive supply voltage for clock generation in a CAN transceiver was fabricated and tested to compensate for this drawback in CAN communication. The system consists of a symmetrical circuit for voltage and current switches, two capacitors, two comparators, and an RS flip-flop. The operational principle is similar to a bistable multivibrator but the oscillation frequency can also be controlled via a bias current and reference voltage. The chip test experimental results show that oscillation frequency and power dissipation are 500 kHz and 5.48 mW, respectively at a supply voltage of 3.3 V. The chip, chip area is 0.021 ㎟, is fabricated with 0.18 μm CMOS technology from SK hynix.
김중술,김영환,유재만 大韓神經精神醫學會 1969 신경정신의학 Vol.8 No.2
This is a study of 35 adolescent psychiatric patients who were treated either at the out-patient clinic or in the psychiatric ward of St. Marry's Hospital, Catholic Medical College for the purpose of diagnostic utility or relevancy of MMPI profiles. Cases are consisted of 13 psychotic and 22 non-psychotic patients aged from 13 to 18 by the end of August, 1969. It was found that the dead line of T-score 70 point was not useful or valid enough tell whether one is psychotic or not with these patients. Almost all of non-psychotic (90.9%) and many of psychotic (54%) patients remained below the dead line although they were suffering from either psychoses or some other psychiatric illnesses. More valid and most helpful was the pattern analysis by experienced clinical psychologists. By this method 61.5% of psychotic patients were classified as psychotic although 70.9% of non-psychotic patients were classified as normal for all their troubles and symptoms. The mean scores and profiles of respective psychotic and non-psychotic patients proved to support above finding saying that, although non-psychotic revealed nothing conspicuous in their mean profiles, psychotic's profile was remarkably eleveted above and near the T-score 60 point line with a pattern suggesting psychotic-like disturbances. It was concluded that MMPI reponses of adolescent psychiatric patients should be interpreted with caution and reservation, not to be given too much weight as a diagnostic tool before more studies are done with it. Hopefully, further investigations will add much and fulfill what it lacks at this time.
자율주행차량의 객체 검출을 위한 카메라-라이다 센서 시각 동기화 방안 연구
최경수(Kyoung Soo Choi),사의환(Yui Hwan Sa),김성진(Sung Jin Kim),강대수(Dae Soo Kang),이정욱(Jung Uck Lee) 한국산학기술학회 2023 한국산학기술학회논문지 Vol.24 No.10
최근 자율주행차량에 관한 다양한 연구가 진행되고 있으며 특히, 인공지능 학습을 위한 GPU와 주변 하드웨어 성능 향상으로 주변 객체 검출을 통해 자율주행차량을 지원하고 있다. 객체 검출을 위한 대표적인 센서는 카메라, 라이다, 레이더가 있으며, 단일 센서의 인지 능력보다 다종 센서의 융합을 통해 검출 정확도를 높이는 것이 높은 신뢰성을 요구하는 자동차 분야에서 선호하는 방식이다. 따라서, 본 논문에서는 카메라와 라이다 센서의 융합 방식을 채택했다. 인공지능 학습 성능을 높이기 위한 요인 중 이기종 센서 간 캘리브레이션과 정밀한 시각 동기화 방식을 통한 객체 검출 성능 검증을 수행했다. 캘리브레이션 방식은 크게 체커보드 기반, 비 목표물 기반, 딥러닝 기반으로 구성되어 있고, 그 중 Zhang’s method 기반 캘리브레이션을 진행하였으며 회전 오차: 0.80°, 이동 오차: 0.02m, 재투영 오차: 7.63 픽셀의 결과를 도출했다. 다종 센서의 시각 동기화는 트리거 신호의 주기와 근적외선 카메라를 통해 확인하였고, 차량의 속도별 투영 이미지를 이용하여 정성적인 성능을 검증했다. Various studies on autonomous vehicles have been conducted. In particular, autonomous vehicles are supported by detecting the surrounding objects by improving the GPU and surrounding H/W performance for artificial intelligence (AI) learning. Representative sensors for object detection include cameras, lidar, and radar, and it is preferred in the automotive field, requiring high reliability to increase detection accuracy through the fusion of multiple sensors rather than the cognitive ability of a single sensor. Therefore, this paper used the convergence method of camera and lidar sensors. Object detection performance was verified through calibration between heterogeneous sensors and precise time synchronization to improve the AI learning performance. The calibration method is largely checkerboard-based, non-target-based, and deep learning-based, of which checkerboard-based calibration was conducted. The following results were derived: rotation error, 0.80°; translation error, 0.02m; reproduction error, 7.63 pixels. Time synchronization of various sensors was confirmed from the period of the trigger signal and the near-infrared camera, and qualitative performance was verified using projection images by the vehicle speed.
Development of a Sensing Module for Standing and Moving Human Body Using a Shutter and PIR Sensor
Ronnie O. Serfa Juan,Jin Su Kim,Yui Hwan Sa,Hi Seok Kim,Hyeong Woo Cha 보안공학연구지원센터 2016 International Journal of Multimedia and Ubiquitous Vol.11 No.7
Advances in technology have led to the development of various sensing devices such as on-body and health-care monitoring that can integrate from different sources and can initiate actions or trigger alarms when needed. This paper developed a sensing module of standing and moving human body using a shutter and a pyroelectric infrared (PIR) sensor. The module consists of Fresnel lens, PIR sensor, interface circuit of the PIR, microcontroller unit (MCU), and light emitting diode (LED) as an alarm indicator. The principle for standing human body is to chop the thermal heat of human body utilized by the camera shutter. The human sensed signal is the controlling part of this algorithm, the detected interruption serves as its function. By merging the whole apparatus including the print circuit board (PCB), the developed product of this proposed sensing module can replace the commercially available detector. Experiment results show that sensing distance is about 7.0 meters with a sensing angles of around 110 degrees at room temperature. In these conditions, the sending ratio is 100% and the module’s power dissipation is rated as 100 mW.