An Identification Method of Distributed Multiple Harmonic Sources Based on Joint Approximate Diagonalization of Eigenmatrices
Zhuo Cheng, Fangwei Xu, Huicong Liu, Huaying Zhang, Kai Liu · 2024
All the existing harmonic source location methods require harmonic impedance parameters and network topology as prior information. However, it is difficult to accurately obtain network parameters under non-fundamental frequency conditions in actual power systems, which makes the above methods difficult to apply in practice. Therefore, this paper proposes a harmonic source identification method without using harmonic impedance parameters. Firstly, all harmonic source signals in the network are solved by the blind source separation algorithm based on multi-point harmonic measurements. Then, separate and recover the harmonic source current on both sides of the point of common coupling (PCC) based on the single harmonic measurements of each load node. Finally, by matching the correlation between the source signal separation results of multi-point measurement data and the separation results derived from single-point measurement data, the position of harmonic sources can be accurately located. The effectiveness and accuracy of the proposed method are verified by simulation cases.