Research On Fault Diagnosis Of Rotating Machinery Gearbox Based On Extreme Learning Machine

Gearbox is the key transmission component of rotating machinery,and its operating state will directly affect the safety and stability of rotating machinery.Once there is a failure,it will affect the efficiency and even lead to the destruction of the aircraft and death.Gearboxes usually require high-load continuous operation,and the operating environment is complex.There are many causes of failure in the environment,as a result the probability of failure is high.Gear and bearing are the components with high failure probability in the gearbox.Therefore,this paper takes the gear and bearing as the research object,and uses the extreme learning machine as the basis to conduct research on feature extraction,intelligent optimization algorithms and pattern recognition.The main contents are summarized as follows:Aiming at the problem that the fault representation of gear in gearbox is not obvious and the accuracy of traditional classification methods are low,a gear fault diagnosis method based on kernel principal component analysis(KPCA)feature extraction and Ant Colony Optimization Extreme Learning Machine(ACA-ELM)is proposed.The time domain and frequency domain features are extracted from the vibration signal of the gear box to form the feature matrix.Because the KPCA method can eliminate redundant features and extract effective feature indicators by fusing multiple feature indicators that characterize the running state of the gear,the KPCA method is used to reduce the dimension of the feature matrix and ensure the performance of the classifier.ELM not only has high learning accuracy,but also has very fast learning speed and good generalization performance.It is gradually used in fault diagnosis and other fields.Aiming at the problem of poor stability caused by the random generation of network parameters in ELM,this paper uses ant colony algorithm to optimize its parameters,constructs ACA-ELM model,and applies this model to the fault diagnosis of gears in rotating machinery gearbox.Compared with ELM,BP,ACA-BP,GA-ELM models,the experimental results show that the method proposed in the paper has good classification recognition effect and stability.Aiming at the problem that fault diagnosis based on single time domain and frequency domain analysis is difficult to accurately identify the fault,a bearing fault classification and identification method based on variational mode decomposition(VMD)and adaptive bat Optimization Extreme Learning Machine(IBA-ELM)is proposed.VMD is an adaptive signal decomposition method,its effectiveness is obtained through simulation analysis,and the key parameters are selected through experiments.VMD is used to decompose the original signal of the bearing in the gearbox,and the permutation entropy and energy entropy features are extracted from the decomposed modal components to form the feature matrix.Bat algorithm,as a new intelligent optimization algorithm,has strong global search ability and fast convergence.Aiming at the problem that it is easy to fall into local extremum in the later stage,an adaptive bat optimization algorithm(IBA)is proposed,which is applied to the network parameter optimization of ELM,and the IBA-ELM model is constructed.And the model is applied to the fault diagnosis of bearing in the gearbox of rotating machinery.At the same time,the fault diagnosis comparative experiment is carried out on BP,ELM,BA-ELM,IBA-ELM.The evaluation of stability and classification accuracy shows that this method is effective for the fault diagnosis of bearings and has certain research significance.