Foreign Object Impact（FOI）is an important event that affects the safe operation of blades, and the online identification of FOI events is conducive to the early warning of huge disasters. To solve the urgent need of a large amount of vibration data during the service of the blade and the need to quickly and accurately identify the FOI event online, a method for the identification of FOI events and the location of the impact moment of the blade under multiple working conditions is proposed, using the joint criterion of Kurtosis and Amplitude Moment（KAM-BTT）. This method is based on the blade tip timing method, and requires only a single sensor. In this paper, a foreign object impact vibration response model considering the blade installation error, blade detuning and sensor installation error is constructed, the FOI simulation is carried out, and a method for obtaining the threshold of foreign object impact identification parameters based on the blade model parameters is proposed. The experimental device for FOI was set up, and the verification of FOI identification at multiple speeds was completed. Experimental results show that the KAM-BTT method proposed in this paper can realize the identification of FOI events within 0.047 s with a single sensor, which provides a research basis for online early warning and data management.

The existence of the aeroengine casing, limited monitoring points, and multi-fault characteristics make obtaining the rotor’s vibration transmission characteristics challenging, resulting in difficulties accurately identifying the rotor unbalance. This paper utilizes a high-frequency composite sensor to monitor the engine’s blade tip clearance (BTC) and extracts unbalanced information from BTC signals for rotor dynamic balancing, while avoiding the need for the once per revolution (OPR) sensor. First, the vibration characteristics of the rotor-blade system under multi-fault conditions are investigated. Then, based on BTC measurement, a none OPR method and an unbalance identification method are proposed, in which the radial vibration of the blade tip in the BTC signals at different speeds is extracted and operated in the time domain to obtain the rotor unbalanced vibration, the signal is reconstructed, and cross-correlation analysis is used to accurately identify the magnitude and phase of the unbalanced signal. Finally, a rotor test bench is utilized for experimental verification. The results reveal that the dynamic balancing method based on the BTC signal can more precisely identify the rotor unbalance than the traditional rotor dynamic balancing method. The application of this technique will effectively improve engine health management and fault prediction.