Machine Learning for Optimizing PV/Wind Smart Grids Using an Improved Spider Wasp Optimizer Algorithm for Learning in Order to Reach SDG7

Abstract

In 2015, the Sustainable Development Goals were established by the United Nations to guide nations in their pursuit
of a future free of environmental degradation. The seventh Sustainable Development Goal is to provide universal
access to affordable, reliable, clean, and sustainable energy. To meet everyday needs in light of changing energy
sources, it is important to modernize power systems. The incorporation of smart grids into the power system
simplifies complex energy networks. Smart grids and converter autonomy are two additional benefits of integrating
Distributed Generation (DG) with RESs like wind and photovoltaic (PV) systems. Nevertheless, smart grids still
have difficulties with frequency fluctuations due to generation and load imbalances, as well as fault breakouts.
While most studies have focused on steady-state models employing machine learning (ML), this study intends to
examine robust performance optimization and talk about future ML applications that might aid with dynamic
behavior during fault breakouts. A smart grid power system model is developed in this study to aid in the
achievement of SDG7. The main improvement is the way the gains are adjusted using meta-heuristic optimization
procedures. This research takes into account the operating system's constraints and uses machine learning based on
the enhanced spider wasp optimizer (SWO). We ran the simulations in MATLAB. The system's use of SWO led to
good outcomes. Subjects—intelligent grid, optimization, green power, AI, errors