Novelty method for the utility harmonic impedance estimation based on Gaussian mixed model

• Published in: IET generation, transmission & distribution transmission & distribution, 2020-07, Vol.14 (13), p.2573-2580
• Main Authors: Tang, Zhirong, Shu, Qin, Xu, Fangwei, Jiang, Yue
• Format: Article
• Language: English
• Subjects: background harmonics; electric impedance; equivalent circuits; existing estimation methods; expectation maximisation iterative method; expectation‐maximisation algorithm; Gaussian distribution; Gaussian mixed model; Gaussian processes; GMM; harmonic analysis; harmonic control; harmonic emission level; harmonic responsibility analysis; iterative methods; Norton equivalent circuit model; novel estimation method; novelty method; parameter estimation; PCC; power system harmonics; probability; Research Article; statistical distributions; utility harmonic impedance estimation
• ISSN: 1751-8687; 1751-8695
• DOI: 10.1049/iet-gtd.2019.1349

The accurate estimation of harmonic impedance on the utility side of a point of common coupling (PCC) is much important to harmonic control, harmonic emission level, and harmonic responsibility analysis. In this study, a novel estimation method for the utility harmonic impedance is proposed, which employs the Gaussian mixed model (GMM) to fit the measured data at a PCC, because the authors find that the statistical distribution of the measured data at a PCC is close to the Gaussian distribution. Also, in theory, GMM can approach the probability distribution of any data. In this method, the Norton equivalent circuit model at a PCC is first expressed as a GMM, and the utility harmonic impedance in the GMM is estimated by expectation maximisation iterative method. Compared to the existing estimation methods, an outstanding advantage of this method is that it is less affected by background harmonics; therefore, higher accuracy can be acquired in this method. The effectiveness and accuracy of this method are verified by simulation experiments and field data.