Determining the harmonic contributions of multiple harmonic sources using data clustering analysis

A new method is proposed in this paper to determine the harmonic contributions of multiple harmonic sources with consideration of the fluctuation of the background harmonic voltage. Affinity propagation and k-means clustering are used for determining each cluster of the harmonic measurement data, and the harmonic contributions of multiple harmonic sources are calculated by partial least-squares regression. The total harmonic contribution index is then used to assess the total harmonic contribution of each harmonic source. The results show that the performance of the proposed method is robust and accuracy in terms of the background harmonic voltage fluctuation, unknown harmonic source and measurement noise.