Study on the estimation of harmonic impedance based on Bayesian optimized Gaussian process regression

System harmonic impedance is a key parameter in the evaluation of harmonic emission level and harmonic responsibility allocation. In order to solve the problem of calculating the harmonic impedance when background harmonics are different, a method of calculating the harmonic impedance based on Bayesian optimized Gaussian process regression (O-GPR) is presented. The harmonic impedance calculation method is based on the harmonic voltage and harmonic current obtained at the point of common coupling (PCC), and the Bayesian optimization of the Gaussian process regression is used to calculate the harmonic impedance of the system. Considering the influence of background harmonics, the harmonic equivalent models with non-Gaussian distribution background harmonics and Gaussian distribution background harmonics are built in the simulation analysis. The accuracy of this method under different background harmonics is analyzed, and its robustness is evaluated by error comparative analysis. Case analysis further verifies the effectiveness of the method. Also, the method is evaluated through indicators such as mean value, standard deviation and root mean square error. It provides ideas for practical engineering applications.