Hybrid Renewable Energy System (HRES) Dataset for Smart Grid Stability under Variable Renewable Fluctuations Using Machine Learning

The current dataset paper presents a comprehensive collection of simulation results generated using HOMER software to support advanced research in grid-integrated hybrid renewable systems. The present dataset captures the operational dynamics of a smart grid framework that integrates solar photovoltaic (PV), wind turbines, and the main grid through a bidirectional converter. The base configuration represents an HRES system supplying a daily electrical demand of 3950 kWh with a peak load of 479.38 kW, as illustrated in the schematic diagram. To study the effect of renewable intermittency on grid stability and optimization performance, two distinct random variability scenarios were modeled: a low variability case (5%) and a high variability case (20%) for both day-to-day and timestep fluctuations. Each variability condition was simulated under multiple operational settings, resulting in a total of 2000 simulation cases per scenario, thereby providing 4000 complete data instances. Each case records key parameters such as power generation, grid interaction, energy conversion efficiency, renewable penetration, and system cost metrics. The present dataset aims to serve as a benchmark resource for researchers focusing on machine learning-based stability analysis, forecasting models, and optimization algorithm validation in renewable energy networks. By incorporating nature-inspired optimization frameworks such as Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), and Adaptive Genetic Algorithms the dataset can be utilized to test dynamic adaptation mechanisms under fluctuating renewable inputs.

Current research work contributes to the ongoing efforts in smart grid stability enhancement, renewable energy integration, and data-driven optimization. It enables users to explore how random variability impacts grid resilience, system losses, and reliability indices, providing a foundation for developing adaptive control strategies and prediction models. The dataset supports both academic and industrial applications related to energy system optimization, hybrid microgrid planning, and renewable resource management, offering a real-world simulation environment that bridges the gap between theoretical modeling and practical implementation.