Fault identification method of micro turbine blade based on the order spectral entropy of bearing vibration

Abstract

Abstract In view of the difficulty in identifying the state of the micro turbine blade, this paper uses the order spectral entropy analysis method to extract the characteristic information of the blade fault based on the measured micro turbine bearing vibration data. Firstly, a micro-turbine test bench was established. The normal temperature stable inlet flow was produced by a blower, the high temperature and pressure unstable airflow was generated by turbojet combustion chamber, and the working inlet mode of a micro-turbine with a wide range of speed changes was simulated. The vibration signal of the bearing was collected by shell drilling, a variety of time-frequency domain feature analysis methods based on vibration signal are difficult to effectively identify blade faults under the combined action of unstable airflow and frequent variable speed. In this paper, the bearing vibration data in time domain is converted to the vibration data in angle domain, and then the order amplitude and entropy were compared and analyzed. The results show that the proposed method can effectively identify the blade fracture and fouling faults under the driving of stable and unstable airflow in the speed range of 0–20 000 r min−1. This method provides a new method for micro turbine blade condition monitoring through bearing vibration data.