Analysis Model And Experimental Verification Of Compressor Blade Vibration Reliability

Compressor blade is such an important part of aero engine that it has to work safely and of high reliability. Affected by various random factors, the vibration characteristics and responses of compressor blade also appear to be random. Researching on the vibration reliability of compressor blades means to study of these random factors and the vibration characteristics and responses of compressor blade by using probability and statistics methods. The routine analysis and reliability analysis are studied in determining mathematical field, but things themselves are fuzzy, vice versa vibration reliability of compressor blades. And study on compressor blade of vibration reliability is in lack of experiments. In this background, researches have been carried out as following:Studies of reliability for compressor blade and existing problems were summarized, and comprehensive vibration reliability of compressor blades was divided into anti-resonance reliability and anti-overload reliability. The analysis model and method of comprehensive vibration reliability for blade were established by response surface-Monte Carlo simulation method. Besides, based on the fuzzy reliability theory, considering the fuzzy of criterion for resonance and amplitude overload, corresponding type of membership functions were introduced, and fuzzy reliability analysis model was established..Experimental study on blade vibration reliability was carried out for that vibration reliability reaching of compressor blade is in lack of experiments. According to distribution of blade feature size and limitation of machining accuracy, test points were designed based on quadratic regression design method. Anti-resonance reliability and anti-overload reliability experiments of blade were conducted and the response surfaces of blade vibration reliability were gained by means of stepwise regression fitting technique. The vibration reliability of blade was studied by Monte Carlo simulation.Model validation was carried out. The reliability numerical analysis results coincided with experimental results, verifying the effectiveness of analysis model and method established. Fuzzy reliability analysis results coincided with experimental better than that of reliability numerical analysis and gained better accuracy for considering the fuzzy of criterion for both resonance and overload..