Forecasting of PV power generation can be categorized into four types based on historical data of PV power output and related meteorological variables. These models include (1) persistence, (2) statistical, (3) machine learning, and (4) hybrid method.

Project SINAG is working on two forecasting models. The first model is statistics-based, called SARIMAX, an extension of SARIMA (Seasonal Autoregressive Integrated Moving Average) with exogenous variables (X). Project SINAG considered weather parameters as the exogenous variables. SARIMAX is a multivariate classical time series method fitted to assess the solar PV output, which exhibits daily seasonality. The other method is called Long Short-Term Memory (LSTM), a deep learning model, and a variety of recurrent neural networks (RNN). LSTM comprises an input layer that receives the raw data, a hidden layer (which may consist of single or several layers) that analyzes the input information, and an output layer that receives the analyzed results and presents the output. This forecasting model aids the non-linear and complex bond between meteorological data under solar PV forecasting.

Reference:

Das, U. K., Tey, K. S., Seyedmahmoudian, M., Mekhilef, S., Idris, M. Y. I., Van Deventer, W., Horan, B., & Stojcevski, A. (2018). Forecasting of photovoltaic power generation and model optimization: A review. Renewable and Sustainable Energy Reviews, 81, 912–928. https://doi.org/10.1016/j.rser.2017.08.017

Return to BLOG POSTS

Return to HOME Page