http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
공성식,문주성,공성식,조봉관 한국철도학회 2023 한국철도학회논문집 Vol.26 No.9
The track circuit divides the rails at regular intervals and detects the occupancy of the train according to the state of current level flowing to the receiver of each section. However, detection of train position in track circuit can only indicate the occupancy status of the train and there are limitations to precise train position detection. As a result, the same signal will be displayed to the subsequent train until the preceding train passes through the section that preceding train is occupying. This will result in inefficient train operation. This study proposes a method to improve the precision of train position detection using a deep learning algorithm that learns and analyzes the voltage change of the track circuit observed during the passage of trains. The proposed method, based on the transmission voltage change of the track circuit, indicates the moving distance from the transmission point by counting the axles that pass the point and uses the deep learning algorithm to reduce the count error. The proposed train position detection method of this study showed better position detection performance than that of the conventional track circuit system.