Harmonic detection and control for power grids have always been major concerns for researchers. With the application of diverse semiconductor materials in power systems, numerous asymmetrical loads arise, resulting in increasingly poor performance of traditional harmonic detection methods. Ensemble empirical mode decomposition (EEMD) provides a new approach for harmonic detection in power systems. Because the harmonic waves in power systems are indeterminate, optimal decomposition results cannot be achieved by means of artificially configured parameters. For such cases, the development of deep neural networks has provided a new solution for harmonic detection. In this study, particle swarm optimization is combined with a deep neural network to establish an adaptive harmonic separation algorithm. By training an adaptive model in this manner, adaptive EEMD can be realized. Moreover, decomposition parameters can be established based on the harmonic content of signals to effectively separate harmonic waves of diverse orders.
INDEX TERMSHarmonic separation; deep neural network; adaptive model; ensemble empirical mode decomposition
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.