<![CDATA[Solar energy is one of the most affordable energy sources and a clean and friendly renewable energy source in the world. Solar energy, which is abundant throughout the world, can be an energy source that is economical to implement. Due to its advantages, the use of solar energy continues to increase throughout the year. One way to convert solar energy into electrical energy is to use a Photovoltaic (PV) or PV Module device. The PV module used as a study is located in the PLN PLTS 7 MWp in Sengkol village. The PV module used as a study is located in the PLN PLTS 7 MWp in Sengkol village. PV energy production forecasting is carried out to assist in managing planning and knowing PV energy production based on meteorological data in an area where PV equipment has not been installed with a user interface in the form of a website. Machine learning is a technique that allows machines to learn directly from examples, data and experience. In contrast to traditional programming approaches where machines are given commands one by one, machine learning can make its own decisions on a problem after it "learns" from the examples or data provided. This research attempts to forecast the energy production produced by photovoltaic devices using the Random Forest Classification algorithm, which is a classification algorithm from machine learning, so that the overall performance results of this modeling are obtained for the data provided. This algorithm works by collecting predictions from a large number of independent decision trees, and then combining these predictions to produce a more accurate and stable final result by involving the use of historical data about temperature, solar radiation, solar radiation, and other environmental characteristics to predict the amount of energy produced by the photovoltaic system at a certain time, so that the overall performance results of this model are obtained for the data provided. The model built was evaluated using a confusion matrix and the results showed that this algorithm achieved an accuracy level of 96%. These results indicate that Random Forest Classification is an effective and reliable method for forecasting photovoltaic energy production.]]>
