From Local Patterns to Global Context: A Multimodal Deep Learning Approach for Complex Power Quality Disturbance Recognition

This paper presents a novel multimodal deep learning framework that advances power quality disturbance recognition by integrating local feature extraction with global contextual understanding. The proposed approach significantly improves classification accuracy and robustness compared with conventional single-modal and shallow learning methods, addressing a critical challenge in modern power systems with high penetration of renewable energy. The outcomes provide a scalable and transferable solution for real-time monitoring and intelligent grid management. With direct relevance to smart grid reliability and decarbonized energy systems, this work demonstrates clear potential for international application and industrial adoption, contributing to more resilient and sustainable electricity infrastructure worldwide.