Design Of Rolling Bearing Fault Diagnosis System Based On Sparse Decomposition Analysis

Rolling bearing,as the key basic component of mechanical system,often works under the condition of harsh environment and heavy load.During the service process,rolling bearings are prone to various failures,causing the mechanical system to fail to work properly or even severely damaged,which greatly affects the safety and economy of the production process.Therefore,research on the health monitoring and early fault diagnosis technology of rolling bearings is of great significance for ensuring safe and reliable operation of mechanical systems.In recent years,the technology of mechanical fault diagnosis has developed rapidly.And the sparse decomposition algorithm has been widely used in this field whose feasibility and effectiveness have been verified.This dissertation aims at the fault diagnosis of rolling bearings.Through analyzing the sparse decomposition theory and its intrinsic properties,and combining with the structure,working principle and dynamic response characteristics of the rolling bearing,the application of sparse decomposition in bearing fault diagnosis is improved,the sparse decomposition efficiency and fault diagnosis accuracy of the algorithm are optimized and a rolling bearing fault diagnosis system based on sparse decomposition is developed.The main contents are as follows:1.The causes and forms of rolling bearing failures are studied,and the vibration mechanisms of rolling bearing under different fault types are analyzed.According to the mechanism of signal generation,the main components contained in the vibration signals are analyzed,and then the simulation signals of various fault types are established.Set up the simulation experiment bench and vibration signal acquisition system on rolling bearing faults,and the vibration signals of different fault types are measured under various working conditions.2.Research existing fault diagnosis methods of rotating machinery,analyze sparse decomposition theory and dictionary building methods.An improved sparse decomposition algorithm is proposed,which improves the efficiency and accuracy of sparse decomposition:In order to improve the efficiency of the sparse decomposition algorithm and the dictionary structure,a method for establishing a grouped wavelet dictionary is proposed,and a fast-orthogonal matching pursuit algorithm is introduced based on the proposed dictionary.In order to improve the efficiency of the sparse decomposition algorithm and improve the dictionary structure,a method for establishing a grouped wavelet dictionary is proposed,and a fast-orthogonal matching pursuit algorithm is proposed based on the dictionary.In order to improve the accuracy of the sparse decomposition algorithm,analyze the influence of the rotational frequency components on the sparse decomposition accuracy and propose improvement measures.The signal-to-noise ratio and the root mean square error are seen as the basis for judgment.Analyze the effects of different types of dictionaries,different atom lengths,and number of matches on the matching results to determine the dictionary,atomic length,and number of matches that are the best match.3.In view of the current poor application of fault diagnosis algorithm engineering,using Python to design a sparse decomposition rolling bearing fault diagnosis software system.The system includes a data file calling and reading module,a fault characteristic frequency calculation module,a data time-frequency domain map display module,a signal processing and time-frequency map display module,and an automatic fault identification and result display module.In order to improve the timeliness of field application of rolling bearing fault diagnosis,a signal pre-processing function is added to this system.Under the condition that the Nyquist sampling theorem is satisfied,the data amount of the signal to be processed is adaptively reduced,the sparse decomposition speed is increased,and the fault feature extraction and rapid diagnosis of rolling bearings in the industrial field are realized.4.The accuracy and effectiveness of the improved sparse decomposition algorithm and fault diagnosis system are verified by simulation analysis and experimental research.