Improved blade tip timing measurements during transient conditions using a state space model

Abstract

Rotor blade condition monitoring is considered an important aspect of turbomachine management, especially with the current trend of systems operating at higher speeds, loads and temperatures. Blade Tip Timing is a non-intrusive method of performing rotor blade condition monitoring during turbomachine operation. The technique uses sensors embedded into the turbomachine casing in conjunction with a shaft encoder. A BTT system calculates the difference between the true and expected Angle of Arrival (AoA) of each blade at each sensor. This difference can be used in many ways to perform diagnostics. When calculating the AoA, the shaft speed within each revolution is often assumed as constant. This assumption is incorrect for transient conditions such as a run-up. The incorrect assumption leads to errors in the calculated AoAs. This article proposes a more accurate method for calculating the AoA. A state space equation for the shaft Instantaneous Angular Speed and Instantaneous Angular Position is derived and solved using a Kalman filter. The proposed method is validated using laboratory and field tests.