FORECASTING PV PRODUCTION USING FUZZY REASONING APPROACH

Donard Shaliu; Arlind Çota; Andi Hida; Raimonda Bualoti; Marialis Çelo

doi:10.51466/JEEIT2491216021sh

Donard Shaliu Faculty of Electrical Engineering, Polytechnic University of Tirana, Albania
Arlind Çota Faculty of Electrical Engineering, Polytechnic University of Tirana, Albania
Andi Hida Faculty of Electrical Engineering, Polytechnic University of Tirana, Albania
Raimonda Bualoti Faculty of Electrical Engineering, Polytechnic University of Tirana, Albania
Marialis Çelo Faculty of Electrical Engineering, Polytechnic University of Tirana, Albania

DOI: https://doi.org/10.51466/JEEIT2491216021sh

Keywords: fuzzy reasoning, influential variables, physical approach, PV generation, statistical approach

Abstract

In recent years, many studies have been done related to the implementation of methods for PV generation forecasting. The PV generation depends on several factors. Since Artificial Intelligence methods have a greater advantage than other methods, in this paper the construction of a PV energy predictive model of using Fuzzy Logic Approach has been considered. In this paper, a fuzzy reasoning approach is used for prediction of the PV generation. In the proposed method, the fuzzy inferences knowledge-based on IF-THEN rule is developed using MATLAB fuzzy software. Using fuzzy logic, the mathematical representation of influential variables on the prediction of solar power is defined. Three variables influencing the prediction of solar power are taken in consideration as fuzzy logic inputs. These variables are: hours of the day, clouds, tilt angle. The detailed analysis of the fuzzy system surfaces shows that the factors taken in consideration are mutually related. A set of different values of inputs are defined and then PV energy production forecast is made applying the fuzzy approach. The constructed rules based in engineering experience accurately represent forecast of PV generation.