Accurate calculation of harmonic impedance is the premise of harmonic control and reasonable division of harmonic responsibility. The existing harmonic impedance estimation methods should meet the conditions that the background harmonic fluctuations are small and the customer harmonic impedance is much larger than the utility harmonic impedance. However,the use of a large number of power electronics in power system makes those conditions are not be satisfied. Therefore,a utility harmonic impedance estimation method based on mutual information and data optimization is proposed. Firstly,the utility harmonic currents are solved by independent component analysis (ICA),and the data segments with small mutual information between the harmonic currents at the point of common coupling (PCC) and the utility harmonic currents are selected. Finally,the harmonic impedance of the screened data segment is estimated using the characteristic of zero covariance between the harmonic current at the PCC and the harmonic voltage on the utility. The effectiveness of the proposed method is tested by simulation and actual case analysis. Compared with existing methods,the proposed method is more accurate and widely applicable when the customer harmonic impedance is not much larger than the utility harmonic impedance.