Enhanced Signal Processing-Based Load Forecasting in Smart Grids Using Artificial Neural Networks and Heuristic Optimization

The incorporation of renewable energy sources into contemporary power systems necessitates accurate management of the demand-supply equilibrium, highlighting the critical role of smart grids.Signal processing techniques play a crucial role in improving load forecasting by refining feature extraction, noise reduction, and classification.This study introduces an innovative signal processing-driven approach for intelligent load forecasting in smart grids, utilizing Artificial Neural Networks (ANN) optimized by the Smart Flower Water Wave Optimization (SFWWO) algorithm.The SFWWO combines Water Wave Optimization (WWO) and Smart Flower Optimization Algorithm (SFOA) to enhance forecasting accuracy and reliability.Additionally, key signal processing techniques, such as feature selection using Motyka and Ruzicka metrics and fusion via Dice Similarity, ensure improved data preprocessing and classification.The ANN_SFWWO model outperforms existing methods, achieving the lowest RMSE (0.225), MSE (0.040), and MAPE (0.770) on the ERCOT Load Data.These findings highlight substantial improvements in energy efficiency, noise-resistant forecasting, and grid stability, underscoring the role of advanced signal processing techniques in optimizing smart grid operations.