Investigation On Vibration Parameters Identification Method Of Rotating Blade Vibration

Blade tip-timing method is the key technologies to study the blade vibration. Tip-timing signal isunder-sampling, so it is very important to investigation on vibration parameters identification methodbased on tip-timing signal. Currently, researchers have proposed a variety tip timing analysis methods,but different conditions apply different analytical methods. This paper summarized the applicability ofeach algorithm and verified the identifying accuracy based on the theory of each algorithm. Finally,developed one set of blade vibration parameter identification software based on the integer order andnon-integer order analysis methods.The main work is as follows:1. Analysis two kinds of data processing methods: Moving average and curve fitting, afterin-depth research, identified as the five-point moving average method. After that, simulation verifiedthe curve fitting result is non-smooth curve while the speed instability.2. Investigation on the basic theory of the integer order tip-timing algorithms, through simulationanalysis prove the speed instability and sensor installation error would affect the identifying accuracyof two-parameter method, but the identifying accuracy of sine fitting and regression method are littleaffected by these situations. Besides that, a non-integer order analysis method is realized, namelyorder tracking method.3. On the basis of research on the theory of algorithms, the simulation analysis results prove thatthe sine fitting and regression method can precisely identify the constant speed synchronous vibrationparameters. For variable-speed sweep simulation data without coupling, three integer order algorithmscan precisely identify the vibration parameters. But for the coupling of simulation data, selecting asingle peak have more accurate identification results than choosing a bimodal, since two-parametermethod is based on the assumption that the single free vibration. For asynchronous vibrationsimulation data, order tracking method can accurately identify the blade vibration parameters.4. Vibration parameter identification software based on above algorithms and simulation analysisis developed using GUI interface tools in Matlab. It has friendly interface, programming norms,complemented by a wealth of mouse and keyboard control, achieved the humane operation.