Time-frequency representation based on time-varying autoregressive model with applications to non-stationary rotor vibration analysis

Abstract

A parametric time-frequency representation is presented based on timevarying autoregressive model (TVAR), followed by applications to non-stationary vibration signal processing. The identification of time-varying model coefficients and the determination of model order, are addressed by means of neural networks and genetic algorithms, respectively. Firstly, a simulated signal which mimic the rotor vibration during run-up stages was processed for a comparative study on TVAR and other non-parametric time-frequency representations such as Short Time Fourier Transform, Continuous Wavelet Transform, Empirical Mode Decomposition, Wigner-Ville Distribution and Choi-Williams Distribution, in terms of their resolutions, accuracy, cross term suppression as well as noise resistance. Secondly, TVAR was applied to analyse non-stationary vibration signals collected from a rotor test rig during run-up stages, with an aim to extract fault symptoms under non-stationary operating conditions. Simulation and experimental results demonstrate that TVAR is an effective solution to non-stationary signal analysis and has strong capability in signal time-frequency feature extraction.

EEG-Based BCI Emotion Recognition: A Survey

“…Model frameworks can categorize classification algorithms [ 56 , 57 ]. The model’s categories may be (1) generative-discriminative, (2) static-dynamic, (3) stable-unstable, and (4) regularized [ 102 , 103 , 104 ].…”

A Novel Time-Frequency Analysis Approach for Nonstationary Time Series Using Multiresolution Wavelet

“…For example, the TVAR models have been successfully applied to the analysis of nonstationary physiological signals including simulation [6], spectral estimation [4], [7], [8], classification diagnosis [9] and synchronization [10]. The coefficients in the TVAR models can also be used to estimate the time-frequency distribution of nonstationary series.…”

“…Therefore, the time-frequency analysis approaches have been applied to investigate the time-frequency characteristics of nonstationary signals and the potential pathologies [1]. Generally, both non-parametric and parametric estimations of time-frequency analysis were adopted in the literatures [2][3][4]. Non-parametric approaches such as Wigner-Villle distribution (WVD), short time Fourier transform (STFT), and continuous wavelet transform (CWT), etc.…”

Review of Parameter Estimation Techniques for Time-Varying Autoregressive Models of Biomedical Signals

“…Although the alternate methods such as WVD yields good resolutions in both time and frequency for systems with single components, however when applied on multicomponent signals they produce a lot of artifacts [8]. Consequently, non-parametric methods are limited by applications and hence are not suitable for broad range of applications.…”