Spectral Angular Speed Estimator (SASE): A Novel Algorithm for the Direct Estimation of Instantaneous Angular Speed Based on Vibration Signatures

In the field of condition monitoring of rotating machines, the estimation of the Instantaneous Angular Speed (IAS) has emerged as a central element. The ability to accurately monitor the rotational speed of a system is a key element of many vibration-based condition monitoring methods. Information on rotational speed is essential to counteract potential speed fluctuations that could otherwise compromise the effectiveness of conventional frequency-based approaches. In practice, the installation of encoders or tachometers may not be possible or cost-effective. Therefore, the challenge of estimating IAS directly from vibration signals has attracted much attention in recent years.

In this study, an innovative method for estimating IAS directly from a vibration signal is presented.

The new method is suitable for non-stationary speeds and can be applied to any speed range and rate of change. It requires only the basic system parameters as input.

The algorithm was verified and validated with simulated and experimental data. These simulations and experimental resultss allowed the verification of the proposed algorithm by calculating the root-mean-square errors between the estimated IAS and the simulated and measured speeds in all scenarios. The observed low maximum error shows the high accuracy of the algorithm.

The strength of this method lies in its ability to correlate several characteristic frequencies of the rotating components of the system, which increases its robustness.