ANNOUNCEMENTS
With the increase in solar power penetration to the grid each year and due to uncertainty and variable nature of solar energy resource, forecasting of solar power is important for economic operation and optimisation of grid control. This study aims to propose an artificial neural network model for solar power generation forecasting. The neural network model performance schemes are based on number of neurons, activation function and optimization algorithm. 80% portion of the dataset has been used to train the model and remaining 20% portion to test the model. The accuracies have been estimated on the basis of mean square error, mean absolute error, mean absolute percentage error and coefficient of determination. The artificial neural network model has been compared with support vector regression and autoregressive integrated moving average models in terms of accuracies and behaviour. The neural network model was found to be more accurate than other models. Data driven performance assessment has been carried out to analyse the performance of photovoltaic solar plant in terms of efficiency and performance ratio using multivariate regression and mathematical descriptive analytics.
Keywords: solar power, forecasting, neural network, efficiency, performance ratio.
