Since it is difficult to identify instantaneous frequency （IF） of noisy response signal accurately by maximum modulus of time-frequency coefficients algorithm， a new method is proposed by combining synchroextracting transform （SET） and maximum modulus of time-frequency coefficients algorithm. At first， the IF is restricted within a range by SET and hence the randomness of searching area selection is avoided. Then， time-frequency ridges and the IF curves with high accuracy are obtained by gradually searching the modulus maxima of time-frequency coefficients in the area restricted by the SET. Two numerical simulations of a mono-component signal and a multi-component signal， a steel cable test with time-varying tension forces and an aluminum cantilever beam test with abrupt mass reduction are used to verify the effectiveness and accuracy of the proposed method. The results demonstrated that the proposed method can effectively extract the IF of time-varying response signal. Compared with the maximum modulus of wavelet coefficients algorithm and synchrosqueezing wavelet transform method， the proposed method behaves better on IF identification and has a good anti-noise property.