The estimation of output power of a photovoltaic system by extreme learning machine

In this thesis, the output power of a PV panel under different operating conditions is estimated via extreme learning machine (ELM). For this purpose, PV panel of 180 W was installed and the open circuit voltage, short circuit current, panel temperature and solar radiation were measured and recorded at regular intervals. 75 experimental data were obtained. The maximum power of the panel is calculated from the open circuit voltage and short circuit current. While the panel temperature and solar radiation are given as input to the regression model of the PV panel formed using ELM, the output of the regression model is taken as the maximum power of the PV panel. Some parameters of ELM such as number of input neurons, activation function were adjusted to provide the best results by trial and error method. The obtained data set is divided into training and test sets. The performance of the method was examined by 5-fold cross-validation method. For this purpose, the data set is divided into 5 equal parts. For this purpose, the data set is divided into 5 equal parts. One of these parts is reserved for testing and the remaining four are used for training the network, and this is done by changing the test set each time. One of these parts is reserved for testing and the remaining four are used for training the network, and this is done by changing the test set each time. A total of 5 times the network is trained and tested with different clusters. The test result of the network is averaged over the sum of the performances of all test functions. The ELM algorithm was implemented using MATLAB software. In addition, different regression models were used to estimate the output power of the PV panel and the results obtained from these models were compared with the results of the ELM. According to the regression results, it is seen that ELM estimates the output power of PV panel with very high accuracy. According to the regression results, it is seen that ELM estimates the output power of PV panel with very high accuracy. In addition, the advantages of ELM as shorter training time compared to other regression models and the fact that the training algorithm is very simple showed that ELM could be used as an effective tool in estimating the output power of PV panels.