Abstract
To overcome the shortcomings that the early fault characteristics of rolling bearing are not easy to be extracted and the identification accuracy is not high enough, a novel collaborative diagnosis method is presented combined with VMD and LSSVM for incipient faults of rolling bearing. First, the basic concept of VMD was introduced in detail, and then, the adaptive selection principle of parameter K in VMD was constructed by instantaneous frequency mean. Furthermore, we used Lagrangian polynomial and Euclidean norm to verify the value of K accurately. Secondly, we proposed a classification algorithm based on PSO-optimized LSSVM. Meanwhile, the flowchart of the classification algorithm of fault modes may be also designed. Third, the experiment shows that the presented algorithm in this paper is effective by using the existing failure data provided by the laboratory of Guangdong Petrochemical Research Institute. Finally, some conclusions and application prospects were discussed.
Highlights
In recent years, the machinery has become more high-speed, intelligentized, and complicated with the development of the modern industrialization
Considering that the mode number K of Variational Mode Decomposition (VMD) needs to be selected according to prior knowledge, improper selection will lead to overdecomposition or underdecomposition so that useful characteristic data cannot be extracted, leading to the problem of low accuracy of fault diagnosis
The leastsquares’ support vector machine (LSSVM) model optimized by particle swarm optimization is combined to carry out fault pattern recognition. e results show the following: (1) e advantage of measuring the value of K by the change of the instantaneous frequency of the signal component after VMD decomposition is more accurate and simple than the previous observation method to judge the value of K, which can avoid overdecomposition and underdecomposition
Summary
The machinery has become more high-speed, intelligentized, and complicated with the development of the modern industrialization. An excellent pattern recognition method of the fault modes has an important influence for the final diagnosis accuracy. Based on this objective, support vector machine (SVM), leastsquares’ support vector machine (LSSVM), BP neural network (BPNN), fuzzy logic (FM), and other methods have been successfully applied in [18,19,20,21,22,23]. Based on the two points mentioned above, an improved fault diagnosis of the bearing will be presented combined with the VMD algorithm based on instantaneous frequency optimization and particle swarm optimization least-squares’ support vector machine in our paper.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
