http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
홍세현,이용우,이훈 충북대학교 산업과학기술연구소 2002 산업과학기술연구 논문집 Vol.16 No.1
Since 1980s, the educational facilities of university are getting explosive, because of enlargement of entering high-level education. Consequently, the new educational buildings of university are constructed, but the management of existing facilities is superannuated. And this problem makes difficult to serve amenity spaces for education. In this thesis, I will find out the efficiently management for the educational facilities of university. I will make the problems clear and suggest the improvement by investigating and analyzing the existing state of the facilities, and the appropriation of the budget, in the there national universities in Chungbuk province.
송영원(Yeong-won Song),박정우(Jeong-woo Park),유진수(Jin-su Yu),홍세현(Se-hyun Hong) 대한기계학회 2022 대한기계학회 춘추학술대회 Vol.2022 No.11
This paper presents an optimal environment construction and efficient approach for rotational machine fault diagnosis. As a research method for this, fault diagnosis is performed at various sampling frequencies based on data acquired by over-specification sensors. Statistical features of data are used for fault diagnosis, and statistical features extracted according to various sampling frequencies and data sample size are used as AI model learning variables. Several ensemble models such as Random Forest and XGBoost were used, and the accuracy of the models was compared and analyzed. As a result, the fault diagnosis performance did not significantly increase after the sampling frequency exceeded 10 times the rotating machine’s maximum frequency. This means that even when a low-performance sensor with a low frequency is used, high performance may be expected. Therefore, this paper is expected to contribute to economically performing fault diagnosis.
송영원(Yeong-won Song),박정우(Jeong-woo Park),유진수(Jin-su Yu),홍세현(Se-hyun Hong) 대한기계학회 2022 대한기계학회 춘추학술대회 Vol.2022 No.11
This paper presents an optimal environment construction and efficient approach for rotational machine fault diagnosis. As a research method for this, fault diagnosis is performed at various sampling frequencies based on data acquired by over-specification sensors. Statistical features of data are used for fault diagnosis, and statistical features extracted according to various sampling frequencies and data sample size are used as AI model learning variables. Several ensemble models such as Random Forest and XGBoost were used, and the accuracy of the models was compared and analyzed. As a result, the fault diagnosis performance did not significantly increase after the sampling frequency exceeded 10 times the rotating machine’s maximum frequency. This means that even when a low-performance sensor with a low frequency is used, high performance may be expected. Therefore, this paper is expected to contribute to economically performing fault diagnosis.