Stochastic Assessment of Renewable Energy Reliability: A Case Study of North Euboea, Greece

The increasing penetration of renewable energy sources (RES) in the energy mix, particularly solar photovoltaic and wind power, poses significant challenges to electricity grid reliability due to their inherent stochastic variability. This study develops a stochastic framework to assess the ability of RES to balance electricity demand, with a focus on storage requirements and reliability implications. Using North Euboea, Greece, as a representative case study, normalized hourly time series of electricity demand, solar irradiance, wind speed, and temperature are analyzed to match per-capita annual energy consumption. Stochastic properties are quantified through climacograms, autocorrelation functions, cross-correlations, and estimation of the Hurst–Kolmogorov exponent, revealing strong long-term persistence in both demand and renewable generation. Results show that, despite annual energy sufficiency, demand is met only 32% of the time for photovoltaics and 44% of the time for wind power in the absence of storage. Introducing moderate storage capacity equivalent to approximately half of the average daily demand (6 kWh per capita) increases reliability to about 70-71%, yet substantial unmet demand and curtailment persist. The weak correlation between wind generation and demand, compared to a moderate correlation identified for photovoltaics, further exacerbates system imbalance. The pronounced long-range dependence of the examined processes implies clustering of deficits and surpluses, significantly increasing reliability risks. The findings demonstrate that achieving high reliability in high-RES systems requires storage and backup capacities far exceeding those implied by average energy balances. Robust energy system planning must therefore explicitly account for stochastic variability, persistence, and demand–supply misalignment when evaluating renewable-dominated power systems.