AI Algorithm Based Green Hydrogen Static Compensator for Harmonic Mitigation, Power Balancing, and Power Factor Correction in Nano-Grids

Abstract

Nano-grids are emerging as a vital solution for sustainable energy distribution, particularly in integrating intermittent renewable energy sources. In this context, this research paper aims to address the challenges faced in Nano-grids through the innovative use of a Green Hydrogen Static Compensator (GHSC). The purpose of this study is to overcome issues related to harmonic pollution, power factor correction, and real power balancing within the Nano-grid system. The method employed involves a two-fold approach. Firstly, a modified instantaneous p-q algorithm is proposed to enhance the effectiveness of GHSC. Secondly, artificial intelligence (AI) algorithms, including deep neural networks, random forest, and XGBoost, are utilized to further refine the system's accuracy. These AI algorithms were trained using the modified p-q method, and the simulation of GHSC was performed using SimScape Toolbox in MATLAB-Simulink. The results of the study clearly demonstrate that the proposed GHSC approach, combined with the AI algorithms, outperforms traditional methods in terms of effectiveness in mitigating harmonic pollution and correcting power factors. However, it is noted that the current implementation lacks flexibility for real-time self-learning. In conclusion, the paper underscores the novelty and potential of the GHSC in conjunction with AI algorithms, while also pointing out areas for future research. Specifically, future efforts may focus on developing an automatic weight updating deep neural network that responds to real-time feedback of THD, power factor, and frequency, enhancing the system's adaptability and efficiency.