Abstract: The nonlinear waveform characteristic of ferroresonance is an important basis for effective detection. However, due to the lack of analytical solution of resonance circuit, it is difficult to give a reasonable detection threshold. In addition, there are faults or disturbances similar to the characteristics of resonance waveform in the actual system, such as arc high-impedance grounding fault, and the accuracy of the existing ferroresonance detection methods is not high. In this paper, the single-phase series ferroresonance and arc high-impedance grounding fault models of neutral point solid grounding system are compared and analyzed,and the time-domain analytical solution of the state equation is obtained by piecewise linearization. A detection method based on the common identification of phase voltage spectrum and phase current mean envelope is designed, and the arc high-impedance grounding fault and power frequency ferromagnetic resonance are further identified by using volt-ampere characteristics. Finally,simulation and field recording data verify the effectiveness of the detection algorithm.