Compressive Sensing Identification Method Of Btt Signals Based On Arbitrary Sensor Arrangement

Since blades are the most significant components of the turbomachinery,it is very important to monitor the vibration status of rotating blades in real time.As a non-contact measurement method,tip timing method has been widely used in vibration monitoring of turbomachinery blades.However,its non-uniform sampling and under-sampling nature also bring difficult to the reconstruction of the signal.Based on the sparsity nature of BTT signal in frequency domain,this paper proposes a method of signal identification based on compressive sensing.The contents of the paper are as follow:1)based on the blade vibration theory of turbomachinery and the basic principle of blade timing system,the sparse reconstruction mathematical model of tip timing signal under arbitrary sensor angle distribution is derived.The application of compressive sensing method in reconstruction of blade tip timing signal and frequency multiplication identification is explored.In order to improve the reconstruction effect and reduce the experimental cost as much as possible,through a large number of numerical experiments of synchronous and non-synchronous signals,the effects of various factors,including signal reconstruction error,number of sampling cycles,frequency resolution and number of sensors,on the identification effect and the optimal range of values were studied.2)In order to reduce the influence of uncertain factors on the results in the experiment,the influence of noise,speed fluctuation,data loss and other random interference with different signal-to-noise ratio on the tip timing signal identification based on the compressive sensing method is analyzed.The influence degree of each random interference on the identification results is obtained by using the global sensitivity analysis method,and the influence is reduced or removed by improving the method.3)Based on the theory of compressive sensing,the optimal sensor layout design and the correlation between a variety of classical sensing performance parameters and BTT signal identification results are explored.An innovative sensor arrangement designing method based on frequency multiplication combination and the characteristics of BTT measurement is proposed.Finally,the specific steps and relevant standards of tip timing signal compressive sensing reconstruction are determined.4)This method is applied to the finite element simulation results of a certain type of aeroengine fan blade and the tip timing data measured by a27-blade rotor,and a new method combining compressive sensing and curve fitting is proposed innovatively.The reconstruction results show that this method can accurately identify the frequency multiplication of the excitations at different rotational speeds,and their corresponding amplitudes are also in good agreement with the strain gauge measurement results.Compared with the traditional curve fitting method and other BTT signal identification methods,this method has the advantages of low dependence on prior information,insensitivity to environmental noise and simultaneous identification of synchronous and asynchronous signals,which verify the method effectiveness and application prospect in engineering.