http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
SCOPUS 문헌 정보 분석을 통한 머신 러닝 활용 BIPV 연구 동향
이제현(Lee Jehyun),유시현(You Sihyun),김창기(Kim Chang Ki),오명찬(Oh Myeongchan),김보영(Kim Boyoung),강용혁(Kang Yong-Heack),김현구(Kim Hyun-Goo) 한국태양에너지학회 2022 한국태양에너지학회 논문집 Vol.42 No.3
With the accelerated development of science and technology across fields, publications in almost all fields have increased exponentially every year. According to the Scopus search, the number of papers related to buildings and solar energy that used machine learning has grown at an average annual rate of 16.1%, exceeding 3000 publications since 2019. Review papers have been published consistently since 2010, demonstrating that they are on a trajectory of initial growth. Because there is a limit to reading and analyzing large quantities of papers every year, we developed a methodology that reads, analyzes, and visualizes published literature information. In this method, we provide a query to the Scopus database to retrieve data and then visualize the number of publications by year, journal, and keyword along with other analyses results. The relationship and frequency analyses results can also be shown among words in titles, keywords, and abstracts. Analysis of research on building-integrated photovoltaics yielded the result that publishing on Energy and Buildings (impact factor [IF] 4.067, modified rank normalized impact factor [mrnIF] top 1.7%) and Solar Energy (IF 4.018, mrnIF top 1.7%) is active. Over the past 10 years, approximately 1–1.5% of machine learning-related research has been published, with a gradual increase.
풍력 발전기 평가를 위한 수집 행렬 데이터 절감 알고리즘 개발
이제현(Jehyun Lee),최정철(Jungchul Choi) 한국신재생에너지학회 2020 신재생에너지 Vol.16 No.1
For mechanical load testing of wind turbines, capture matrix is constructed for various range of wind speeds according to the international standard IEC 61400-13. The conventional method wastes considerable amount of data by its invalid data policy -segment data into 10 minutes then remove invalid ones. Previously, we have suggested an alternative way to save the total amount of data to build a capture matrix, but the efficient selection of data has been still under question. The paper introduces optimization algorithms to construct capture matrix with less data. Heuristic algorithm (simple stacking and lowest frequency first), population method (particle swarm optimization) and Q-Learning accompanied with epsilon-greedy exploration are compared. All algorithms show better performance than the conventional way, where the distribution of enhancement was quite diverse. Among the algorithms, the best performance was achieved by heuristic method (lowest frequency first), and similarly by particle swarm optimization: Approximately 28% of data reduction in average and more than 40% in maximum. On the other hand, unexpectedly, the worst performance was achieved by Q-Learning, which was a promising candidate at the beginning. This study is helpful for not only wind turbine evaluation particularly the viewpoint of cost, but also understanding nature of wind speed data.