Short-term forecasting of global solar irradiance with incomplete data
This work addresses forecasting for renewable energy and agriculture planning, but it is incremental as it applies existing methods to a specific dataset with missing data handling.
The paper tackles short-term forecasting of global solar irradiance with incomplete data by introducing a pipeline that includes data imputation and uses data-driven models like ARIMA and neural networks, with results showing neural networks outperform ARIMA, especially LSTM in cloudy environments.
Accurate mechanisms for forecasting solar irradiance and insolation provide important information for the planning of renewable energy and agriculture projects as well as for environmental and socio-economical studies. This research introduces a pipeline for the one-day ahead forecasting of solar irradiance and insolation that only requires solar irradiance historical data for training. Furthermore, our approach is able to deal with missing data since it includes a data imputation state. In the prediction stage, we consider four data-driven approaches: Autoregressive Integrated Moving Average (ARIMA), Single Layer Feed Forward Network (SL-FNN), Multiple Layer Feed Forward Network (FL-FNN), and Long Short-Term Memory (LSTM). The experiments are performed in a real-world dataset collected with 12 Automatic Weather Stations (AWS) located in the Nariño - Colombia. The results show that the neural network-based models outperform ARIMA in most cases. Furthermore, LSTM exhibits better performance in cloudy environments (where more randomness is expected).