**6. Conclusions**

In the sector of PV power forecasting, machine learning techniques within artificial intelligence offer a lot of potential. The main benefit of these approaches is their ability to handle complex problems and take into consideration a large number of input factors, However, it is worth noting that selecting an optimum number of input variables is beneficial for successful machine learning, since large datasets can be difficult to analyze and interpret. As a result, the PCA approach is critical, as it allows for faster computations and storage space savings, as well as the removal of redundant variables, multicollinearity, and noise.

Finally, the comparison of machine learning approaches for PV power forecasting will aid energy suppliers in identifying the best algorithms for effectively and safely handling PV-integrated power.
