태양광 설비 및 기후 데이터를 활용한 태양광 발전량 예측 모델 개발

• Published in: 한국CDE학회 논문집 Vol. 29; no. 1; pp. 33 - 41
• Main Authors: 우태경(Tae-Kyeong Woo), 소민섭(Min-Seop So), 강신영(Shin-Young Kang), 신종호(Jong-Ho Shin)
• Format: Journal Article
• Language: Korean
• Published: (사)한국CDE학회 01.03.2024; 한국CDE학회
• Subjects: 기계공학; Transformer model; Photovoltaic power generation; Power generation prediction; Artificial intelligence
• ISSN: 2508-4003; 2508-402X
• DOI: 10.7315/CDE.2024.033

Faced with ongoing resource depletion and climate catastrophes due to global warming, there is an increasing interest in reducing carbon emissions and the need for sustainable renewable energy. To address these issues, the expansion of renewable energy usage is imperative. Among various sources, solar power generation is the most widely used source in renewable energy production and occupies the largest proportion. This study aims to propose a methodology for predicting solar power generation using environmental and power generation data. Considering the characteristics of solar power generation, which are influenced by the continuity of time, a Transformer model specialized in time series data was developed. This model demonstrates per- formance superiority compared to other existing AI models. The proposed model shows promis- ing results and is expected to be utilized for advanced operation and management of power plants. KCI Citation Count: 5