Novel Grid Adaptive Harmonic Discrimination Based on Improved CEEMDAN

Accurate discrimination of grid harmonics has been a hot issue for researchers. Ensemble Empirical Mode Decomposition (EEMD) introduces a novel approach to tackle this challenge. However, the intricacies and uncertainties surrounding harmonics in modern power systems often result in rigid parameter configurations, limited computational efficiency, and suboptimal harmonic identification accuracy with EEMD. To overcome these limitations, an Enhanced Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) is proposed as a viable alternative for harmonic discrimination. This study integrates ICEEMDAN with a collaborative architecture between cloud and end devices to devise an adaptive harmonic discrimination technique. Experimental findings demonstrate the method's effectiveness in mitigating modal aliasing issues inherent in EMD, offering flexibility for deployment across terminal detection devices, thereby establishing its efficacy and superiority.