A Novel Approach for the Detection and Classification of Power Quality Disturbances Using a Hybrid Deep Learning Model

• The hybrid CNN-BiLSTM model achieves 99.2% classification accuracy for power quality disturbances, outperforming traditional methods. • The model is robust to noise and varying sampling rates, making it suitable for real-time monitoring in smart grids. • The proposed approach reduces computational complexity while maintaining high accuracy, enabling deployment on edge devices. • The study provides a comprehensive dataset and benchmark for future research in power quality disturbance classification.