English title

---

Parameter estimation of photovoltaic cell model based on measured characteristics is difficult due to the nonlinear behavior of the diode in the model and its exponential relation. Existence of two diodes in double-exponential model makes this estimation more difficult. A hybrid optimization algorithm is implemented based on explorative characteristic of invasive weed algorithm, probabilistic models of distribution estimation and utilizing dispersion capacities of a mixed Gaussian–Cauchy distribution to estimate parameters of the single and double diode models of photovoltaic cells as an optimization problem. The fitness function is defined as the root mean square error between current-voltage curves of the model output and the measured points. The proposed method is implemented for a practical photovoltaic cell based on measured characteristic and the obtained results are compared with other 8 optimization algorithms. The statistical comparison of the difference between current-voltage curve of the optimized circuit model and the measurements verifies more suitable performance of the implemented algorithm compared with others.

---

This study is aimed to apply three machine learning techniques including Random Forest (RF), Support Vector Machine (SVM), and Multivariate Adaptively Regression Spline (MARS) in both short-term and long-term electric load forecasting problems and compare their performance. Last hour load, temperature, and solar altitude angle of the present hour with holidays were considered as inputs. Three different criteria including root mean square error, mean absolute error, and coefficient of determination were used to evaluate the performance of prediction methods. These methods are all applied to practical electric load demand data obtained from a sub-transmission substation in Hamedan using R programming language. The temperature data are collected from the nearest meteorological weather station and the hourly solar altitude angle for the whole year is accurately calculated with astronomical equations for the studied location. The results show that the implemented methods provide acceptable forecasts and RF and SVM models exhibit superb results and provide more accurate forecasts in short-term load and long-term load forecasting respectively.