Research On Fault Diagnosis Method Of Desulfurization Pump Coupling Based On RF-CNN

Damage or failure of rolling bearings in production equipment can lead to a chain reaction and may cause serious consequences.Desulphurization pump as a common chemical production process for the transport of corrosive liquid equipment,its use of harsh environment,so the desulphurization pump coupling in the rolling bearing condition monitoring and fault diagnosis to timely and accurate,so as to ensure the safe and reliable operation of machinery and equipment,to avoid the occurrence of unexpected accidents and reduce maintenance costs,etc.has a vital role.At present,more and more enterprises are exploring sensor and cloud-based rolling bearing condition monitoring and fault diagnosis systems to provide comprehensive and accurate rolling bearing condition assessment and fault diagnosis services through technologies such as big data analysis and machine learning.Effective diagnosis of impending faults of rolling bearings can improve the operational efficiency and productivity of mechanical equipment.This paper takes rolling bearings as the research object,and explores the rolling bearing fault diagnosis technology by analysing the vibration mechanism,failure form and common fault diagnosis methods of rolling bearings,with random forest algorithm as the research basis.The main research content of this paper includes the following aspects:(1)Explains the mechanical structure and vibration mechanism of rolling bearings,analyses the failure forms of common rolling bearings,the causes of failure and their performance,and introduces the common diagnostic methods for rolling bearing fault diagnosis from two aspects: mechanism analysis and intelligent diagnosis.(2)Introduces the basic theoretical knowledge of random forest,describes in detail the generation of decision tree algorithm,and explains the three performance indicators for measuring random forest.The specific steps of constructing a random forest network model are described,and the advantages and disadvantages of random forest are analysed.Experiments are carried out to verify the effectiveness of the random forest network model in the application of rolling bearing fault diagnosis for second classification.(3)A detailed description of the convolutional neural network model and the components of the model,and a description of the training process of the convolutional neural network model.On the basis of this study,a combined network model combining the convolutional neural network model and the random forest algorithm is proposed,and the combined network model is applied to the study of rolling bearing fault diagnosis multi-classification,and through experiments,a comparative analysis with the traditional convolutional neural network model is conducted.