http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Semantic Segmentation을 이용한 구름 분석 기술 기반, 일사량 및 태양광 발전량 예측 기술
이다훈(Da-Hun Lee),고디모데(Di-Mo-De Go),김현우(Hyun-Woo Kim) 대한전자공학회 2020 대한전자공학회 학술대회 Vol.2020 No.11
As interest in solar power generation increases, solar power generation facilities and power generation are also increasing. Estimating the amount of power generated in energy generation is essential. However, in the case of solar power generation, it is difficult to predict the amount of radiation or power generation because it is greatly affected by the atmospheric environment. In this paper, to solve this problem, SKY images are acquired through an IP camera equipped with fisheye lens. After that, the image is analyzed using the segmentation method to obtain cloud information. We propose a system that predicts the amount of radiation and generation by using the obtained information as an input value of the LSTM algorithm. In particular, it was confirmed that when cloud information was obtained by the segmentation method, the accuracy was higher than that of the previously used methods. From the data obtained after that, it was confirmed that the error rate was within 10% in predicting the amount of radiation and power generation through LSTM.