Enhanced firefly algorithm inspired by cell communication mechanism and genetic algorithm for short-term electricity load forecasting

Electricity load forecasting plays a pivotal role in energy management systems, enabling efficient resource allocation and optimal power grid operation. This paper proposes a hybrid approach for short-term electricity load forecasting by integrating a neural network model with the enhanced firefly algorithm (EFA), inspired by cell communication mechanisms, and a genetic algorithm (GA). The proposed methodology leverages the neural network's ability to capture complex patterns from historical load data while utilizing metaheuristic optimization techniques to enhance forecasting accuracy. The EFA, designed to improve exploration and exploitation capabilities, refines parameter selection within the optimization process, while the GA further fine-tunes neural network parameters to enhance model performance. Extensive experimentation on Nigeria’s TCN-NCC electricity load dataset demonstrates the effectiveness of this approach. The hybrid CCMFA-GA-ANN model achieves a mean absolute percentage error (MAPE) of 1.07%, outperforming other benchmark models such as CCMFA (1.26%), BA (1.22%), FA (1.21%), and GA (1.19%). The model also achieves the lowest mean absolute error (MAE) of 48.00 and the highest forecast efficiency of 0.52. Additionally, the Pearson correlation coefficient of 0.99969 and a coefficient of determination (R²) of 0.99999 indicate a strong agreement between actual and predicted values. With a rapid convergence time of 2.321 seconds, the hybrid approach ensures computational efficiency, making it suitable for real-time forecasting applications.These results highlight the significant improvements in forecasting accuracy achieved by the proposed approach compared to conventional methods. The model’s high accuracy and efficiency make it a valuable tool for energy management systems, aiding decision-making in grid operations, demand-side management, and infrastructure planning.