Probabilistic analysis and scenario generation of load uncertainty using statistical methods

With the growing penetration of renewable energy, the randomness and volatility of load demand have become key challenges for stable power system operation. To handle the uncertainty of load forecasting, this paper proposes a data-driven energy storage scheduling approach based on scenario generation and intelligent optimization. First, historical load data are statistically analyzed to identify probabilistic characteristics. Monte Carlo sampling is applied to generate multiple load uncertainty scenarios. The Whale Optimization Algorithm (WOA) is then employed to optimize the energy storage scheduling strategy for each scenario, improving system adaptability to varying load conditions. Finally, boxplot analysis is used to statistically evaluate the distribution, stability, and robustness of the optimization results. The results indicate that the proposed method can effectively mitigate the impact of load fluctuation, enhance the stability of energy storage scheduling, and improve overall system performance.