Rolling Bearing Fault Diagnosis Technology Based On Resonance Method

As a key component of rotating machinery,rolling bearings are frequently used in various fields.Rolling bearing runs in harsh condition for a long time,which is easy to fail due to various faults.At the same time,it causes equipment shutdown and even catastrophic accidents.Thus,it results in huge losses.Rolling bearing and other components constitute a coupled nonlinear system,which makes the vibration signal of bearing have the characteristics of randomness and non-stationary.In addition,in the early stage of rolling bearing fault,the weak vibration characteristic information is submerged in the strong noise background.Therefore,the bearing fault is difficult to extract.These factors seriously affect the fault identification and diagnosis of rolling bearing.Therefore,the nonlinear system resonance method is used to study the fault characteristic extraction and fault diagnosis of rolling bearing.The specific contents of this paper are as follows:(1)The strong resonance phenomenon may appear in the response of nonlinear system by periodic excitation.Therefore,based on the resonance phenomenon,a fault diagnosis method of rolling bearing is proposed.The system resonance method enhances the fault characteristics and is easy to identify.At the same time,based on the improved quality factor,the resonance induced by external excitation is determined.It also improves the accuracy of rolling bearing fault diagnosis.In addition,this method can not only enhance the fault characteristics,but also suppress other frequency components in the spectrum.Both numerical simulation and experimental verification illustrate the correctness and effectiveness of the proposed method.The bearing fault diagnosis is successfully realized by the system resonance method.(2)Considering the existence of engineering background noise,based on adaptive stochastic resonance and coherence resonance,this paper studies the fault diagnosis and fault severity evaluation of rolling bearing under strong noise background.In fact,if we do not distinguish coherence resonance and stochastic resonance,it may lead to misjudgment.Therefore,in order to avoid misjudgment,the coherence resonance theory is introduced to distinguish it from stochastic resonance,so as to realize the judgment of bearing fault.By numerical simulation and experimental verification,the necessity of distinction is illustrated.At the same time,the method improves the accuracy of bearing fault diagnosis.In addition,the transition from adaptive coherence resonance to stochastic resonance corresponds tothe change of bearing from a healthy state to a fault state.Therefore,with the help of this nonlinear system response transformation,the severity of early rolling bearing fault is evaluated in the strong noise background.Based on the analysis of experimental signals with different severity,it shows that the method can reflect the severity of the early bearing fault.(3)In the actual engineering background,there is weak noise or even no noise background.In the previous research,it is found that the transition from self-induced system resonance to system resonance corresponds to the change of bearing from a healthy state to a fault state.Therefore,this paper evaluates the severity of early rolling bearing fault in the background of weak noise by this nonlinear system response transformation.The analysis of bearing fault experimental signals with different severity shows that the method successfully realizes the early fault severity evaluation in the weak noise background.In addition,compared with the square root amplitude and effective value,the method has high sensitivity in early fault severity evaluation.(4)In order to realize the extraction of rolling bearing fault under the condition of variable rotating speed,this paper proposes a piecewise re-scaled aperiodic resonance method.Firstly,the aperiodic resonance is studied by using several different types of nonlinear frequency modulation signals.Then,the weak signal is enhanced and extracted by combining with short-time Fourier transform.Finally,the validity of the method is verified based on the experiment signal with variable speed.The results show that the method can not only enhance and extract the time-varying characteristic frequency of rolling bearing fault,but also remove noise and superharmonic interference.(5)We summarized the work and made a prospect for the related researches.This thesis has 91 figures,12 tables,155 references.