Research On The Performance Monitoring And Fault Diagnosis System Of High-speed Gearbox

As one of the key components of the mechanical equipment,high-speed gearboxis widely used in the machinery industry.With the constant development to high-speed,overloading and high efficiency of the mechanical equipment,the performance of high-speed gearbox has greater impact on the production practice.Therefore,it has very important value of engineering application to vigorously develop and research onthe effective equipment condition monitoring and fault diagnosis technology.It can improve the efficiency of equipment operation,reduce the cost of equipment maintenance and management,and improve the economic benefit and safety equipment performance.In this thesis,as the research object,high-speed gearbox was researched on its condition monitoring and fault diagnosis method to construct a high-speed gearbox performance monitoring and fault diagnosis system.The main research contents of this thesis include:1)According to the problem that gearbox vibration signal contains a large amount of status information which is difficult to extract effectively,wavelet packet transform was used to de-noise and extract energy feature of the original vibration signal.BP neural network was used to achieve the fault patternrecognition.Due to the problem that neural network was at slow convergence speed and easy to fall into local optimal value,simple and easy particle swarm algorithm was put forward to optimize weights and bias of BP neural network.Through the simulation experiment of the gearbox fault,the method was compared and analyzed with the basic particle swarm algorithm and genetic algorithm respectively.The results showed that this method can effectively improve the efficiency and accuracy of classification.2)In view of the shortcomings of the current nonlinear and non-stationary signal processing methods,the adaptive morphology based on variational mode decomposition(VMD)was proposed to effectively extract the characteristic frequency of the signal.VMD was used to decompose the target signal into a finite number of modal signals firstly.Then the modal signal was extracted based on mutual information method,and summed to reconstruct a signal.The morphologic filter was used to reduce noise from the reconstructed signal,and extracted the frequency of the rolling bearing fault feature.Aiming at the problem of morphologic structural selection and the inherent statistical deviation,the particle swarm optimization was used to adoptively optimize the improved generalized morphological filter.Through the digital simulation experiment and rolling bearing fault experiment,the method was compared with the adaptive morphology based on EMD and envelope demodulation.The results showed that this method can extract the fault characteristic frequency effectively.3)Built high-speed gearbox performance monitoring platform.Several common structures of gearbox test-bed were compared and analyzed.The test-bed of opened power was chosen to be built.Advantech data acquisition card and industrial control computer was used as the platform to build a data acquisition system.The construction of the performance monitoring platform provides the basis for performance test and the experimental research of high-speed gearbox.4)Based on the research of high-speed gearbox performance monitoring platform and gearbox fault diagnosis method,hybrid programming of MATLAB and LabVIEWwas used to design the software part of performance monitoring and fault diagnosis system.Through comparing the common method of hybrid programming by MATLAB and Lab VIEW,method based on COM component technology was selected for designing the software.According to the demand of performance monitoring and fault diagnosis,the module of data acquisition,vibration signal analysis,fault feature extraction,fault intelligent diagnosis and axis trajectory analysis were designed.The software could realize the function of on-line monitoring and off-line analysis which would provide a platform for practical engineering application.5)The performance monitoring and fault diagnosis system were verified by the running experiment of high-speed gearbox and the analysis of fault diagnosis examples.The no-load and load experiment were curried on the high-speed gearbox test-bed,and the results verified the validity of the data acquisition system.The transmission efficiency of high-speed gearbox was researched with the test data.The calculation module of transmission efficiency was analyzed and compared with the test data,which provided a basis for the research on the power loss of high-speed gearbox.Through the fault diagnosis case of high-speed gearbox,the fault diagnosis module was verified.The result showed that the fault diagnosis module could identify the fault types of the high-speed gearbox effectively.