Enhancing solar radiation forecasting using machine learning algorithms

Abstrak

With the increasing amount of photovoltaic (PV) generation, accurate solar radiation forecasting is essential to the safe operation of power systems. This work examines many machines learning (ML) techniques that use both exogenous and endogenous inputs to forecast sun radiation. In order to find pertinent input parameters and their values based on previous observations, the forecasting models’ performance is assessed using metrics like mean absolute error (MAE), mean squared error (MSE), R-squared (R2), and root mean squared error (RMSE). Accurate power output forecasting is becoming more and more necessary as the need to switch to renewable energy sources (RES) like solar and wind power grows. There is a clear demand for more reliable solutions because current models frequently struggle with temporal complexity and noise. A revolutionary deep learning-based technique designed especially for green energy power forecasting was developed in response. The study uses time series smoothing and the autoregressive integrated moving average (ARIMA) model for casing in order to create a solid basis for analysis and modeling that is free of noise and outliers. The proposed method aims to address the limitations of existing forecasting methods and promote the creation of more accurate and reliable forecasts in the field of renewable energy.