Research On Rolling Bearing Fault Diagnosis System Based On Symplectic Geometric Mode Decomposition

With the widely use of rotaing machinery in the industrial field,in order to ensure the safe operation of the equipment,the fault diagnosis and monitoring of the important parts of rotating machinery,especially the rolling bearing,becomes more and more important.Due to the sophisticated working environment of rolling bearing,the vibration signals collected have a large background noise,a low signal-to-noise ratio,and difficulty in identifying fault characteristics.In response to this problem,this article conducts research on the rolling bearing fault diagnosis system based on the symplectic geometry mode decomposition,and developed a rolling bearing fault diagnosis system.The main research contents of this article are followed:(1)Aiming at the problem of difficulty extraction of rolling bearing fault signal features,a improve symplectic geometry mode decomposition(ISGMD)method is proposed.Firstly,the dynamic time warping(DTW)method is applied to the recombination of SGMD similar components,which solves the problem that the method is not clear in the recombination of SGMD components;secondly,waveform matching extension method is used as the preprocessing method of SGMD algorithm to solve the problem of end effect of SGMD decomposition results;finally,the validity of the proposed method in the diagnosis of rolling bearings is verified by simulation signals and experimental bearing signals.(2)Aiming at the problem that the traditional symplectic geometry mode decomposition method relies on empirical formulas for the selection of embedding dimensions,a parameteroptimized symplectic geometry mode decomposition(OSGMD)method is proposed.First of all,the comprehensive evaluation target value function of a modal aliasing index,overdecomposition index and kurtosis index is introduced to select the optimal embedding dimension;secondly,the ramanujan periodic transform(RPT)method is used to enhance weak fault characteristics and obtain more feature information;finally,the results of simulation and experimental analysis show that the method of combining OSGMD and RPT has an excellent anti-model mixing and anti-noise interference ability,and can effectively extract the fault characteristics in the strong noise background.(3)Aiming at the problem that the single-channel acquisition signal fault information may be insufficient in the traditional rolling bearing fault diagnosis,SGMD is extended to the complex domain and a complex symplectic geometry mode decomposition(CSGMD)method is proposed.Firstly,the two-channel vibration signal is synthesized into complex signal and decomposed and reconstructed by CSGMD method to obtain corresponding positive and negative frequency components;secondly,in order to further fuse fault information,full vector spectrum technology is used to process the signal,and the positive and negative frequency components of the signal are fully vector fused,taking into account the differences in fault characteristics between the two channels;finally,the effectiveness of the proposed method in multi-sensor information fusion is verified using the Xijiao public dataset.(4)For the requirements of online diagnosis of rolling bearings,a rolling bearing fault diagnosis system is developed based on the C# programming language.Firstly,the overall plan design of the fault diagnostic system platform to determine the model of the sensor and data collection card used in the experiment;then,the software collection and post-processing algorithm module is designed and developed,including the development of vibration signal processing functions such as real-time signal acquisition,data management,EMD,VMD,LMD,SGMD,extraction of time-frequency characteristic parameters such as steepness,mean value,frequency of center of gravity and fault sound field cloud diagram;finally,the system is deployed on the experimental platform and the software-related functional test is performed through the experimental signal.The results showed that the developed rolling bearing fault diagnosis software system can analyze and process fault signal and meet the design and application requirements of the system.