首页|Reports from Faculty of Information Engineering Highlight Recent Research in Sup port Vector Machines (Fault diagnosis method using MVMD signal reconstruction an d MMDE-GNDO feature extraction and MPA-SVM)
Reports from Faculty of Information Engineering Highlight Recent Research in Sup port Vector Machines (Fault diagnosis method using MVMD signal reconstruction an d MMDE-GNDO feature extraction and MPA-SVM)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Data detailed on support vector machin es have been presented. According to news originating from Quzhou, People's Repu blic of China, by NewsRx correspondents, research stated, "To achieve a comprehe nsive and accurate diagnosis of faults in rolling bearings, a method for diagnos ing rolling bearing faults has been proposed. This method is based on Multivaria te Variational Mode Decomposition (MVMD) signal reconstruction, Multivariate Mul tiscale Dispersion Entropy (MMDE)-Generalized Normal Distribution Optimization ( GNDO), and Marine predators' algorithm-based optimization support vector machine (MPA-SVM)." Our news correspondents obtained a quote from the research from Faculty of Infor mation Engineering: "Firstly, by using a joint evaluation function (energy*|corr elation coefficient|), the multi-channel vibration signals of rolling bearings a fter MVMD decomposition are denoised and reconstructed. Afterward, MMDE is appli ed to fuse the information from the reconstructed signal and construct a high-di mensional fault feature set. Following that, GNDO is used to select features and extract a subset of low-dimensional features that are sensitive and easy to cla ssify. Finally, MPA is used to realize the adaptive selection of important param eters in the SVM classifier. Fault diagnosis experiments are carried out using d atasets provided by the Case Western Reserve University (CWRU) and Paderborn Uni versity (PU). The MVMD signal reconstruction method can effectively filter out t he noise components of each channel. MMDEGNDO can availably mine multi-channel fault features and eliminate redundant (or interference) items."
Faculty of Information EngineeringQuzh ouPeople's Republic of ChinaAsiaMachine LearningSupport Vector Machines
NETL
NSTL
