Accurate parameter identification is crucial for developing reliable, discrete simulation models, designing enhanced condition diagnosis schemes, and preventing system failures. Non-invasive methods without external excitation are preferred for parameter estimation. This article proposes a non-invasive parameter identification approach based on Adaptive Approximate Bayesian Computation with Sequential Monte Carlo Sampler for DAB DC-DC converters. The proposed approach can simultaneously identify both circuit and control parameters, including passive, parasitic, and control loop elements, using only the input/output variable under steady state and transient conditions. The results demonstrate that the proposed method can accurately and efficiently identify the parameters of a DAB converter. The algorithm's flexibility makes it applicable to other parameter identification problems and optimization platforms, such as rectifiers and filters.