With the growing amount of renewable energy sources being connected to the grid, issues concerning the power quality are becoming increasingly important. This is due to the distributed nature of these power sources and to their higher variability when compared to more traditional sources. For photovoltaic sources, total harmonic distortion of current is one of the main power quality parameter closely related to solar irradiance. This work explores this relationship and employs computational intelligence techniques in order to predict current distortion values. Preliminary results confirm the possibility to capture these relationships and predict occurrence of power quality issues ahead of time.