Abstract: Due to the advantages of high-cost performance, accuracy and digitization, digital twin has become an advanced technology for fault trending and predictive maintenance of power electronic converters. However, the current digital twin models of power electronic converters fail to consider the fractional characteristics of actual inductors and capacitors, posing a challenge for accurate prediction and fault diagnosis. To address this problem, this paper proposes a prediction-correction digital twin model for power electronic circuits based on fractional calculus. The inductance (L) and capacitance (C) are identified at different fractional orders using a digital twin parameter identification method based on particle swarm optimization (PSO), and the equivalent series resistance (ESR) is calculated. Compared to existing methods, the proposed method not only improves the identification accuracy of the actual inductance and capacitance, but also identifies the fractional parameters at different fractional orders and capacitances. Finally, implementation of continuous conduction mode (CCM) Boost converter with different inductances, capacitances and fractional orders is presented. Besides, different working conditions and identification numbers are considered. The experimental results verify the effectiveness of the proposed model and method.