A novel amplitude-FRF based SDOF resonator parameter extraction method

Abstract

Single degree-of-freedom (SDOF) resonators are commonly used in modeling of dynamical systems. The SDOF resonator parameter extraction methods utilize the complex form of the frequency response function (FRF), which may not be available. The half-power bandwidth does not require the complex FRF, but uses only three points of the amplitude response, making it ineffective when the FRF data is noisy or incomplete. In this study, a linear regression based SDOF resonator parameter extraction method is presented. The method applies weighted least squares to the manipulated resonator FRF. The performance of the proposed method is compared with least-squares regression based on inverse-squared FRF and the half-power bandwidth method, using noise and nonlinearity added FRF data and experimental data. It is concluded that the proposed method outweighs other methods using amplitude FRF.