A robust solution for power grid management using a hybrid deterministic and probabilistic model for short term load forecasting

Short term load forecasting (STLF) is crucial for maintaining energy balance and optimizing operational efficiency in electricity distribution networks. While various methods exist for load estimation, this study introduces a novel hybrid deterministic-probabilistic algorithm. This study introduces a novel hybrid forecasting model SAMM (Seasonal-Adjusted Mycielski-Markov) which integrates three distinct components: seasonal adjustment to isolate periodic patterns, the deterministic Mycielski method to identify historical recurrences, and the probabilistic Markov chain model to handle uncertainty in future states. Using 3 years of hourly electricity consumption data from several cities in western Turkey, the proposed model is validated and shows promising outcomes compared to traditional methods. The novel SAMM algorithm outperforms conventional approaches, providing a robust solution for power grid management. Evaluation metrics such as RMSE, MAE, MAPE, and R² highlight the effectiveness of the proposed algorithm: 1.79, 0.7925, 4.07%, and 0.9791, respectively. The paper contributes a novel methodology to the field of short-term load forecasting, offering a framework open to future enhancements and integration with other innovative smart grid applications for sustainable cities.