A Study on the Wind Power Forecasting Model Using Transfer Learning Approach

Recently, wind power plants that generate energy with electricity are attracting a lot of attention thanks to their smaller installation area and cheaper generation costs. In generation, it is important predict the amount generated because system would be unstable due uncertainty in supply. However, difficult accurately varies several causes, such as speed, direction, temperature, etc. this study, we deal mid-term (one day ahead) forecasting problem data-driven approach. particular, intended solve newly completed generator makes very lack data on past generation. To end, deep learning based transfer model was proposed compared other models, without Light Gradient Boosting Machine (LGBM). As per experimental results, when applied similar complex same region, confirmed low predictive performance constructed could supplemented.