Modelling Solar Power Generation in Europe Using Support Vector Regression

In this work we modelled the solar power generation in Europe using support vector regression. The modelling also performed for three countries: Germany, France and United Kingdom. In the modelling the effect of using different kernels was observed. Radial kernel outperformed linear and polynomial kernels. Also using radial kernel, effect of modelling one country separately and using all Europe data together was investigated. It was observed that for Germany and United Kingdom using Europe data together with the countries data produced lower average error and root mean square error values. On the other hand, for France using only France data gave a better modelling compared with using France and Europe data together.