Abstract: To cope with the frequently occurred subsynchronous oscillation (abbr. SSO) in large-scale grid-connected wind power system, it is necessary to develop a method to identify and detect SSO quickly and accurately. During the occurrence of SSO there are such features as time-varying characteristics and uncertainty and these bring the challenge to the realtime monitoring of SSO. In allusion to this problem, firstly, a solution based on leading in intrinsic time-scale decomposition (abbr. ITD) improved by algebraic estimation was proposed, the proposed solution did not need any priori information and its performance was not affected by the frequency constitution of the SSO. Secondly, by use of synthetic signals, simulation results of Electro-Magnetic Transient Program (abbr. EMTP) and the measured data of SSO the research of comprehensive comparison was conducted, and the research results showed that in the aspects of dynamic performance of signal check and the accuracy of parameter estimation the proposed solution achieved good results. Finally, by means of hardware-in-the-loop test the feasibility of the proposed solution is verified.