Rolling Bearing Fault Diagnosis Base On Adaptive Stochastic Resonance

Rolling bearing fault diagnosis has been studied to prevent severe mechanical equipment damage and ensure the long-term safe operation of mechanical equipment.Unexpected machinery failure causes bodily injury and economic loss.A typical process of bearing fault diagnosis can be described as follows:1)the vibration signals are acquired from fault bearings,2)envelope demodulation and frequency analysis methods are utilized to obtain the envelope spectrum,and 3)the existence and type of faults are confirmed.However,extracting fault features is difficult because fault signals are always corrupted by heavy background noise.This thesis studies an adaptive stochastic resonance method and its application in rolling fault bearing.Stochastic resonance has the characteristic of enhancing weak characteristic signal.In this thesis,a USSSR method is used in fault dignosis.The unique secondary-filtering ability can enhance weak signal and improve SNR,Stochastic resonance is a nonlinear parameterized filter,and the output signal relies on the system parameters for the deterministic input signal.The most commonly used index for parameter tuning in the stochastic resonance procedure is the SNR.However,using the SNR index to evaluate the denoising effect of stochastic resonance quantitatively is insufficient when the target signal frequency cannot be estimated accurately.To address this issue,six different indexes,namely,power spectral kurtosis of the stochastic resonance output signal,correlation coefficient between the stochastic resonance output and the original signal,peak SNR,structural similarity,root mean square error,and smoothness,are constructed in this study to measure the stochastic resonance output quantitatively.These six quantitative indexes are fused into a SQI via a back propagation neural network to guide the adaptive parameter selection of the stochastic resonance procedure.The index fusion procedure reduces the instability of each index and thus improves the robustness of parameter tming.In addition,GA is utilized to quickly select the optimal stochastic resonance parameters.To sum up,this thesis studies the adaptive stochastic resonance method based on SQI and its application in fault diagnosis of rolling bearing.Above studies are based on the simulation and experimental verification.The proposed method is compared with the traditional method,and with advantages of high accuracy,less time-consuming,easy to be implemented.At the same time,the practicability and effectiveness of the proposed method in the fault diagnosis of rolling bearing are further verified.