http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
고백시와 SNS 글쓰기에 나타난 저자와 대중의 관계 연구 - 실비아 플라스의 「나사로 부인」을 중심으로
변세희 ( Sehee Byun ) 현대영미어문학회 2018 현대영미어문학회 춘계학술대회 발표논문집 Vol.2018 No.-
This paper explores various shared traits between confessional poems of the 1960s and personal SNS writing culture currently established online and aims to demonstrate how we might read them both as part of the larger conversation about modern-day culture of confessional and autobiographical writing. This paper mainly examines how Plath’s poem, “Lady Lazarus” functions as highly performative works of art intended to elicit a desired effect from readers. This paper interprets Plath’s ability to make meaningful connections with her readership as remarkably similar to SNS writers’ relationships with their respective current on-line communities.
RTC를 이용한 전기이륜차 보조배터리 방전 방지 알고리즘
변세희(Sehee Byun),백남철(Namchul Paik),곽인재(Injae Kwak),이준우(Junwoo Lee),김상훈(Sanghoon Kim) 한국자동차공학회 2023 한국자동차공학회 학술대회 및 전시회 Vol.2023 No.11
In this paper, we propose an auxiliary battery discharge prevention algorithm for semi-permanent use of auxiliary batteries for electric two-wheelers. The market for electric two-wheelers is growing due to the recent increase in delivery demand due to the COVID-19 and eco-friendly policies in India and Southeast Asian countries. In order to be competitive in the growing electric two-wheeler market, it is necessary not only to improve driving performance but also to develop various additional functions, so the use of auxiliary batteries in electric two-wheelers is essential and important. To prevent the auxiliary battery from being discharged, when the vehicle is not in use, the VCU wakes up periodically to check the auxiliary battery voltage. If necessary, it turns on the high voltage system to charge the auxiliary battery using the main batteries as a charging source. For this purpose, the ADC and RTC alarm wakeup functions of the VCU are utilized. In addition, data was calibrated considering the parasitic current of the entire system and the capacity and charge/discharge characteristics of the auxiliary battery. Finally, the effectiveness of the algorithm was validated through repeated charging and discharging tests in an actual vehicle.
김상훈(Sanghoon Kim),곽인재(Injae Kwak),변세희(Sehee Byun),김대우(Daewoo Kim) 한국자동차공학회 2022 한국자동차공학회 학술대회 및 전시회 Vol.2022 No.11
This study relates to the design of a vehicle control system for controlling an electric two-wheeler system with a drive device operating at a high voltage. In recent years, Micro-mobility has been gaining traction as the new means of mobility of the future, and it is expected that in the future the adoption of short-range use or non-owned shared vehicles in the mobility market will expand. Among the micro-mobility, electric two-wheelers are eco-friendly and economical, and in addition to this, vehicle sharing must be accompanied by a battery exchange system and data-based services. In this case, the electric two-wheeler system generally consists of 5 sub-systems : the powertrain system, the energy management system, the cooling system, the vehicle control system, and the connectivity system. In order to control such an electric two-wheeler system, it is necessary to design an appropriate system platform for the element system to be controlled and a design process for each controller. In this paper, we have focused on designing a control system architecture and control method to ensure that the control is done appropriately from the perspective that the vehicle control system, which is the top-level controller, views the subsystem as the object of control. The sub-control modules of the vehicle control system were constituted and the functional elements of each module were defined to enable organic cooperative control between the control objects through architectural design and verification according to the V-Process. This is expected to improve the productivity and quality of more diverse and complex electric two-wheeler systems.
전기이륜차 커넥티드 차량 제어 시스템 개발에 대한 연구
김상훈(Sanghoon Kim),곽인재(Injae Kwak),변세희(Sehee Byun),김대우(Daewoo Kim) 한국자동차공학회 2023 한국자동차공학회 학술대회 및 전시회 Vol.2023 No.11
With the continuous development of Internet of Things technology, the field of connected cars, which is its application field, is also attracting attention. A connected car is a vehicle that has connectivity with an external network using communication technologies such as LTE and 5G, and is an active vehicle control system that collects data, generates information, and exchanges control data with each other for real-time processing. In the case of electric two-wheeled vehicles, it is also applied as a major technology, and it uses the LTE / BLE communication method through the modem and Bluetooth device mounted on the VCU, which is the top-level controller of the vehicle, and uses the web server, smartphone app, Bluetooth keyfob. In addition, it includes vehicle control functions such as theft detection notification, keyfob command reception, and remote blocking of vehicle operation, as well as vehicle speed, mileage, and warning lights, which are basic functions of the passenger car cluster, and the status information of the vehicle is stored on the server through the modem, as well as the GPS-based vehicle movement location record and driver usage information such as the users vehicle driving pattern. In this way, it is necessary to supply power at all times through an auxiliary battery for functional operation in a situation where the driver is not on board, which requires close cooperation with the transition of the operating mode of the vehicle. In this paper, we propose a software architecture design for connectivity services of electric two-wheelers and present the possibility through actual vehicle verification tests.
온 디바이스 AI 기반 주행 환경 인식을 이용한 퍼스널 모빌리티 안전 제어 연구
강수혁(Soohyeok Kang),김근수(Geunsu Kim),박규도(Gyudo Park),노형태(Hyeongtae Noh),변세희(Sehee Byun) 한국자동차공학회 2022 한국자동차공학회 학술대회 및 전시회 Vol.2022 No.11
In this paper, we propose a method for recognizing the current driving environment and limiting the moving speed. For environment recognition, an on-device AI system using a smartphone’s camera and gps are configured. The recognized result is transmitted to mobility through bluetooth communication. In mobility, the logic that controls driving speed through torque control was designed according to the recognition result. Since this system has limited computing resources, it is necessary to use model compression techniques such as quantization or pruning to reduce the computational load while maintaining the performance of the AI model. So the trained model was quantized into types float16, int8, and full int8. The inference performance of quantized model was evaluated and applied on the CPU, GPU, and NPU of Samsung’s Galaxy S22 Ultra smartphone. Also, we tested the proposed method in the actual mobility system using the MOBILGO of the Hyundai KEFICO’s electric two-wheeled vehicle system.