<![CDATA[Solar power plants have several advantages, namely continuous production, reduced electricity demand, low maintenance of Photovoltaic (PV) and PV life of more than 30 years, so that the use of solar panels can be optimized by using PV power output predictions. The goal is to determine the PV power output for the future. PV power output prediction can use Artificial Neural Network (ANN). In this study, a comparison was made of PV power output predictions using the Cascade Forward Neural Network (CFNN) and Feed Forward Neural Network (FFNN) using the Levenberg-Marquard Algorithm as the activation function of the PV power output prediction learning process. The magnitude of the error is calculated using the Mean Square Error (MSE). From the results of research using the Cascade Forward Neural Network (CFNN) method with the Levenberg-Marquard algorithm, it is obtained that the MSE results are better at a learning rate of 0.1 with an MSE of 0.0042% while for the Feed Forward Neural network (FFNN) it also uses the Levenberg- Marquard obtained MSE results of 0.007% with a learning rate of 0.05. The research results show that CFNN gives the best MSE value, so that the smallest MSE value is used as a reference in energy management systems to predict PV power output.]]>
Copyrights © 2022
