Predicting the Power Output of a Grid-Connected Solar Panel Using Multi-Input Support Vector Regression

As the penetration of photovoltaic power is increasing, utilities are concerned about its impact on distribution grid. Due to the variable nature of solar power, predicting the power output of solar panel installation is important for its optimal use. This paper proposes a new method for forecasting the power output from a solar panel using multi input Support Vector Regression model. The performance has been analysed and compared with Analytical PV power forecasting model. Both the models are simulated and performance evaluation is done using MATLAB. Mean Absolute Percentage Error and Mean Absolute Error are used to assess forecasting models.