In the aftermath of power system faults

electrical waveforms may experience significant distortion and reduce signal-to-noise ratios. Traditional frequency measurement methods employed by phasor measurement units (PMUs) can result in frequency measurement errors ranging from several hertz to tens of hertz. Such erroneous measurements lack reference value and may even jeopardize the normal operation of the power grid and equipment. To address this issue

this paper proposes a PMU electromagnetic transient frequency correction method based on an improved dual complex band-pass filter model. Initially

it defines two types of information entropy based on the frequency and power spectra of post-fault waveforms and introduces an information entropy-based method to quantify the degree of waveform distortion. Building upon the flexibility and speed of the dual complex band-pass filter algorithm

this paper presents an enhanced frequency correction calculation model that enables real-time adaptive correction of frequency measurement results during electromagnetic transients. Furthermore

this paper analyzes the characteristics of information entropy during quasi-steady and fault periods

proposing a method to identify post-fault electromagnetic transient processes

which provides initiation and termination criteria for the correction method. Simulation tests and experimental evaluations using field fault recording data demonstrate that the proposed method reduces the maximum measurement offset of post-fault frequency and frequency rate of change to one-tenth of that of the reference methods.