P2　Enhancing the accuracy of deep learning models for solar irradiation prediction

抄録

Improving the accuracy of deep learning models for solar irradiation prediction is crucial for the optimization of home energy management system (HEMS). This study proposes a new deep learning model to better perform open-loop prediction. The convolutional neural network (CNN) and long short-term memory (LSTM) models are introduced to enhance the accuracy of the model for different weather conditions and locations. The prediction results and performance of the model will be evaluated using solar irradiation data from regions such as Sapporo, Tokyo and Osaka in different weathers, and compared with LSTM based model and CNN model. The results indicate that the hybrid CNN-LSTM model outperforms the other two models overall and shows better performance under stable condition.

ポスター