A Practicable Copula-Based Approach for Power Forecasting of Small-Scale Photovoltaic Systems
Sadegh Rajabalizadeh, Seyed Masoud Moghaddas Tafreshi
Abstract
Photovoltaic (PV) systems are well known as one of the most significant sources of distributed generations in the world. Nowadays, the usage of small-scale PVs by households in residential homes has increased noticeably. The output power of PVs is associated with uncertainty, so there is a need for a practicable method to forecast the output power that can be used by households. The current article presents a feasible method to predict the power of small-scale PVs in a day-ahead pattern. This applicable model does not have the complexities of other forecasting methods and can be applied by households in general. On the other hand, power produced by PVs depends on meteorological characteristics. Meanwhile, one important point to be noted is the fact that in many models the dependency of the PVs output power on meteorological characteristics is taken into account, but the correlation among these meteorological features is neglected. Therefore, in this article, the copula is used in order to model stochastic correlation structure among meteorological characteristics, such as ambient temperature, wind speed, and solar irradiance. Subsequently, by using a mathematical model for obtaining the output power of the PVs in parallel with utilizing the Monte Carlo method the prediction is performed. Here, to study the proposed method, a PV unit of a residential home in Swanson region in New Zealand is considered. The simulation results demonstrated high accuracy of the proposed method in diverse meteorological conditions.