SUMMARY
System Identification is an important and often complex process in many areas of engineering. This process is not easy when parameters of the analysed system vary with time. In such cases classical methods fail to identify parameters properly. The work demonstrated in this paper deals with identification of natural frequencies of time-variant systems. The paper presents the application of the Time-Variant Frequency Response Function for this analysis. Calculation procedure requires division of output spectrum by input spectrum which often leads to division by close to zero values, and that leads to infinite (or undefined) values of resulting transfer function. Additional processing is required for interpretation. The major focus and challenge relate to ridge extraction of the above time-frequency characteristics. The methods presented in the paper are illustrated using experimental multi-degree-of-freedom system. The results show that the proposed method captures correctly the dynamics of the analysed time-variant systems.