Short Term PV Power Estimation by means of Extreme Learning Machine and Support Vector Machine

PV power estimation is a basic stage in solar system implementations such as missing data forecasting, power flow control and fault detection. Generally, intelligent methods are preferred to estimate PV power since they are more compatible with nonlinear problems. This paper presents a comparison of Extreme Learning Machine (ELM) and Support Vector Machine (SVM) for PV power estimation. Temperature and power measurements obtained from a PV plant in Sanhurfa are employed as data set to test the methods. The results have been analyzed for sunny, mid-cloudy and cloudy days in summer. The results reveal that ELM has a better performance than SVM in terms of the accuracy of PV power estimation.