Optimal design and performance assessment of a data-driven hybrid sustainable energy management framework for grid-tied urban power networks

Frequent power outages, climate-related uncertainties, and the increasing complexity of urban power systems have heightened the demand for dependable and economically viable renewable energy solutions. This study develops a grid-tied hybrid sustainable energy structure for an urban power network in Guangxi province, Southern China, combining Photovoltaic and Wind generation, Battery storage, and utility grid support. A Firefighter Optimization (FFO) MPPT controller improves boost converter performance,optimize power extraction from PV systems under variable atmospheric conditions. Quantitative simulation results show that proposed MPPT approach reduces convergence time up to 47.27% and improves dynamic recovery performance up to 46.15% compared to conventional GWO and PSO-based controllers. In parallel, an FFO-optimized artificial neural network (ANN) based energy management strategy (EMS) is developed to coordinate power flow among sustainable sources, battery storage, and the grid. The EMS utilizes PV power, wind generation, load demand, and battery SOC to generate control signals for battery charging–discharging and grid power exchange. Techno-economic analysis demonstrates that designed configuration significantly reduces net present cost and levelized cost of energy compared to conventional grid-dependent operation. Sensitivity analysis reveals a ± 15% change in load demand results in 33.41% and 39.18% variation in NPC and annual operating cost. ±15% change in solar GHI results in a 31.18% variation in the RF and ±15% variation in wind speed leads to change 11.5885% and 3.55% in RF and IRR respectively. Overall, the findings demonstrated that adequate control and optimal system configuration in sustainable micro-grids improve system reliability and long-term solutions in urban power network.