Abstract: For the existence of low frequency oscillation phenomena in the interconnection power systems,a new algorithm for detecting the dominant low frequency oscillation modes was proposed to avoid the disadvantages of previous methods that could not identify the dominant modes accurately and could not adaptively reveal the oscillation modes’ time-varying performance.According to oscillation signal characteristics,damped sinusoid model was selected to represent atomic library.Using the power angle trajectory,the modal parameters were identified from the atomic library by atomic decomposition method and atomic decomposition energy entropy was calculated during the identification process.By comparing the energy entropies,the dominant inertial modes were identified.This method,unrestricted from the system order,has a strong ability of data processing and can reveal the complex dynamic characteristics and nonlinear effects among the modes.So the method can be applied to low frequency oscillation on-line analysis.The results of two simulation cases show that this algorithm can accurately detect the dominant modes and the simulation results are identical to that of the normal form theory,and the accuracy and effectiveness of the proposed method were validated.