Performance Improvement of Grid-Connected PV-Wind Hybrid Systems Using Adaptive Neuro-Fuzzy Inference System and Fuzzy FOPID Advanced Control With OPAL-RT

Abstract

This paper presented enhancing hybrid energy systems, specifically those combining photovoltaic (PV) and wind turbine sources, linked to the electrical grid with the use of an Adaptive Neuro-Fuzzy Inference System (ANFIS) control unit during implementation. Our study focuses on the comparative performance of the ANFIS controller against traditional Fuzzy Fraction-Order Proportional-Integral-Derivative, (FOPID) and Fuzzy Proportional-Integral, (PI) controllers. Hybrid energy systems present unique challenges due to renewable energy sources’ intermittent and non-linear nature. Conventional controllers, such as Fuzzy FOPID and Fuzzy PI, often struggle to manage these complexities effectively. The ANFIS controller, nevertheless, blends fuzzy logic’s qualitative reasoning with neural networks’ capacity for adaptive learning, offering a more robust and flexible solution. Through extensive simulations and real-world testing, we demonstrate that the ANFIS controller significantly outperforms both Fuzzy FOPID and Fuzzy PI controllers in key performance metrics. These include improved voltage regulation, lower total harmonic distortion (THD), and enhanced overall system stability and efficiency under varying load and environmental conditions. The findings highlight ANFIS’s potential as a better hybrid energy system control method, enabling more dependable and effective grid integration of renewable energy sources. This research contributes to advancing smart grid technologies and promoting sustainable and resilient energy infrastructure. Using the OPAL-RT 4512 platform, this paper generates a thorough real-time simulation and investigation of a hybrid PV/wind energy system. Real-time integration of intricate MATLAB/Simulink models is made possible by the OPAL-RT 4512, which makes it easier to accurately simulate real-world operational situations. At the beginning time the values and percentage modified ANFIS and Fuzzy FOPID give good controller compared Fuzzy PI, at time (0 to 0.4 sec) Fuzzy PI decrease in most results values percentage 20% compared two controller ANFIS and Fuzzy FOPID.