A Novel Forecasting Model for Solar Power Generation by a Deep Learning Framework with Data Preprocessing and Postprocessing

Abstract

Photovoltaic power has become one of the most popular energy due to environmental factors. However, solar power generation has brought many challenges for power system operations. To optimize safety and reduce costs of power system operations, an accurate and reliable solar power forecasting model is significance. This study proposes a deep learning method to improve the performance of short-term solar power forecasting, which includes data preprocessing, feature engineering, Kernel Principal Component Analysis, Gated Recurrent Unit Network training mode based on time of the day classification, and post processing with error correction. Both historical solar power, solar irradiance, and Numerical Weather Prediction (NWP) data, such as temperature, irradiance, rainfall, wind speed, air pressure, humidity, are considered as input dataset in this work. As a case study, the measured solar power data from ten solar sites in Taiwan are forecasted for the next day PV power outputs with one-hour resolution. The error index such as Normalized Root Mean Squared Error (NRMSE), Normalized Mean Absolute Percent Error (NMAPE) are chosen to evaluate the performance of forecasting models. Compared with other benchmark models including ANN, LSTM, XGBoost, and single GRU, the experimental results by the proposed forecasting model show its high performance. Furthermore, the proposed model also demonstrates the importance of data preprocessing and post processing based on error correction.