ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술대회 Kalman Filter-based Adaptive Forecasting of PV Power Output
Cited 0 time in scopus Download 5 time Share share facebook twitter linkedin kakaostory
저자
이현용, 김낙우, 이준기, 이병탁
발행일
202010
출처
International Conference on Information and Communication Technology Convergence (ICTC) 2020, pp.1342-1347
DOI
https://dx.doi.org/10.1109/ICTC49870.2020.9289532
협약과제
20ZK1100, 호남권 지역산업 기반 ICT 융합기술 고도화 지원사업, 이길행
초록
In building a forecasting model, weather conditions are typical inputs. However, PV faults happen unexpectedly and affect PV power output significantly. Considering that reliable forecasting is still required in such cases, a forecasting model needs to be adaptive to such PV faults until PV faults are fixed. In pursuing that adaptive forecasting, we utilize Kalman filter together with forecast history. Observing recent forecast history relevant to the current forecast, we try to adjust forecast output by reflecting the observations through Kalman filter. Through experiments using the real-world data, we show that our approach quickly realizes adaptive forecasting just when relevant forecast history is available while the effectiveness of the re-training approach (i.e., one representative existing approach) is limited by available data about PV faults. Our approach (the re-training approach) increases MAPE from 15.46% to 20.27% (33.3%) as the amount of drop of PV power output increases from 0% to 30%.
키워드
adaptation, deep learning, Forecast, Kalman filter, PV
KSP 제안 키워드
Available data, Filter-based, PV power, Real-world data, deep learning(DL), forecasting model, kalman filter, power output, re-training, weather conditions