AUTHOR=Shanmugam S. , Sharmila A. TITLE=An intelligent adaptive neuro-fuzzy based control for multiport DC-AC converter with differential power processing converter for hybrid renewable power generation systems JOURNAL=Frontiers in Energy Research VOLUME=12 YEAR=2024 URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2024.1471265 DOI=10.3389/fenrg.2024.1471265 ISSN=2296-598X ABSTRACT=

The increasing demand for renewable energy sources necessitates the development of sophisticated control systems that can seamlessly integrate and manage multiple power sources. This research introduces an advanced intelligent adaptive neuro fuzzy-based control (IANFC) for multiport DC-AC converters with differential power processing (DPP) converters, tailored for customized hybrid renewable power generation systems (HRPGS). The system aims to optimize HRPGS performance and efficiency through neuro-fuzzy control techniques. When integrating different DC power sources, such solar panels and wind turbines, into AC loads or the grid, multiport DC-AC converters are essential. These converters reduce the amount of power conversion steps, which improves the system’s overall efficiency and scalability. Complementary DPP converters process only the differential power, thereby significantly reducing total power consumption and conversion losses. The IANFC framework combines fuzzy logic reasoning, based on rules, with neural network adaptive learning capabilities. This hybrid control method effectively manages the nonlinear and dynamic behavior of HRPGS, ensuring reliable performance under varying load demands and environmental conditions. The controller dynamically adjusts the converter’s operating point to ensure optimal power flow and system stability. Simulation findings using MATLAB/Simulink verify the efficacy of the suggested IANFC system. Under various operational situations, key performance measures like response time, stability, and system efficiency are examined. As evidenced by the data, system performance has significantly improved as compared to traditional control techniques. The proposed system demonstrates an efficiency of 99.45% and achieves stability in just 0.02 s. Compared to conventional algorithms, this approach shows superior performance across multiple metrics.