The Investigation Of Methods Of Harmonic Excitation Identification In Operational Modal Analysis

Operational Modal Analysis (OMA) is widely used in several fields in recent years. Operational Modal Analysis is mainly to deal with the engineering problems from the input signals that we can hardly measure, but the output signals are recorded easily. Its advantage is we don't need the input signal. It is useful in situation where it is difficult or impossible to artificially excite the structure. The presence of harmonic components in the measured responses is unavoidable in many application of Operational Modal Analysis. This is especially true when measuring on mechanical structures containing rotating or reciprocating parts.The presence of harmonic components has a very bad effect on the operational modal analysis. Harmonic components may be mistaken for being structure modes; it might bias the estimation of the structural modes (natural frequency, modal damping, mode shape).The paper summarizes several methods of harmonic identification. The method of density of probability of signal and the method of kurtosis are particularly analysed. The signal used in these methods is single degree of freedom, so filters are needed in the process of harmonic identification. The paper mainly includes the following :All the methods of digital filer design are summarized in detail. According to the technical specifications, a Butterworth digital band-pass filter is designed. Based on corresponding MATLAB program, the filter is tested by simulation signal produced in MTALAB. The characters of the Morlet wavelet transform in time-frequency domain are also introduced in this paper. Based on Morlet wave, a filter is designed to separate all components in the signal to be identified. The filter is tested by simulation signal produced in MTALAB and is validated to be feasible.The differences of probability density curve of different variables are displayed, the method of probability density is developed according to literature. The method is validated by simulation signal and GARTEUR plane model signal, and the simulation result shows that the method is feasible. The meaning and definition of kurtosis is introduced, and the identification method based on kurtosis is developed. The method is validated by simulation signal and GARTEUR plane model signal, and the simulation result shows that the method is feasible.The validation test is operated with girder modal. Using the two harmonic identification methods, the harmonic component aroused by a motor is identified. The results show the two harmonic identification methods are feasible and effective.