An improved fitting-base keyphasor-free blade tip timing method and experimental validation

Abstract

Blade tip timing (BTT) is a popular non-intrusive blade vibration online measurement technology. Most BTT systems rely on a once-per-revolution (OPR) or keyphasor sensor to provide accurate references. However, such a sensor could be difficult to install due to limited space in field tests. The keyphasor recovery method is the dominant method for keyphasorfree measurements, and several algorithms have been proposed in previous studies, while their accuracy has not been rigorously verified by comparison. In this paper, an improved method based on the linear fitting idea is proposed, and a test rig that can provide the benchmark of the blade tip displacement is designed. The displacement measurement accuracy of the proposed method in this paper is experimentally compared with three other keyphasor-free methods. The results demonstrated that the displacement measurement results of the proposed method are closest to the benchmark under lowspeed acceleration, synchronous vibration and non-synchronous vibration conditions, and the root mean square (RMS) of the measurement error is less than 0.05 mm. This indicates that the BTT is capable of completing long time online health monitoring without keyphasor sensors.