| Centrifugal compressor has many advantages,such as compact structure,high-pressure ratio,and wide scope of stable operation.It has been widely applied in aerospace,energy,chemical,and metallurgical sectors.When the internal flow field of centrifugal compressor is in the unsteady flow state,the centrifugal impeller can easily produce fatigue failure due to the strong fluid excitation.Therefore,the operation safety of impellers in the unsteady flow state has become an urgent problem.In order to study the relative characteristics of blade vibration and noise in unsteady flow state of centrifugal compressor,this paper studies the unsteady flow characteristics of internal flow field in a centrifugal compressor test bench,mainly from two aspects of the inherent unsteady flow and the unsteady flow.The related vibration and noise signals are tested and the signal characteristics are extracted for the two unsteady flow states of rotor-stator interaction and rotating stall respectively.The specific research contents are as follows:(1)In view of the unsteady airflow excited force and high amplitude pressure pulsation caused by the rotor-stator interaction in centrifugal compressor,the recognition of the signal characteristics of rotor-stator interaction is studied.Firstly,the experimental speed conditions of rotor-stator interaction are determined through the operational modal simulation calculation of the impeller and the drawing of the impeller interferogram.Then,strain signals and pressure pulsation signals are collected during steady conditions and sweep frequency process respectively.Finally,by comparing the analysis results of strain signals and pressure pulsation signals,the existence of the interference characteristic frequencies under the simulated rotating speed conditions is verified,and the strain signals are proved to be more suitable for identifying the characteristics of the interference signals.(2)Pressure pulsation signals test for centrifugal compressor test bench is carried out under the rotating stall state,and stall characteristic frequencies are detected based on pressure pulsation signals of airflow in compressor.By studying stall mechanism and summarizing stall signal characteristics,the rotating stall recognition and prediction method based on pressure pulsation signals monitoring is put forward,and it will lay the foundation for subsequent online monitoring of rotating stall for compressor.(3)In view of the fact that the stall characteristic frequency in low frequency of pressure pulsation signal spectrum sometimes is not very obvious when compressor is in stall,the relevant enhancement method for weak stall signals is studied,including envelope demodulation and stochastic resonance algorithms.Finally,the feasibility of the proposed enhancement method is verified by the pressure pulsation data of weak stall signals.(4)According to the proposed prediction method for rotating stall recognition based on pressure pulsation signals monitoring,the stall monitoring system based on envelope demodulation and stochastic resonance algorithms is established by using the LabVIEW software,and the relevant verification of it is carried out.And it is finally applied to the state monitoring and maintenance of compressors in the actual working conditions,which achieves the purpose of stall early warning. |
PDF Full Text Request