Defective rolling bearing response is often characterized by the presence of periodic impulses. However, the in-situ sampled vibration signal is ordinarily mixed with ambient noises and easy to be interfered even submerged. The hybrid approach combining the second generation wavelet denoising with morphological filter is presented. The raw signal is purified using the second generation wavelet. The difference between the closing and opening operator is employed as the morphology filter to extract the periodicity impulsive features from the purified signal and the defect information is easily to be extracted from the corresponding frequency spectrum. The proposed approach is evaluated by simulations and vibration signals from defective bearings with inner race fault, outer race fault, rolling element fault and compound faults, respectively. Results show that the ambient noises can be fully restrained and the defect information of the above defective bearings is well extracted, which demonstrates that the approach is feasible and effective for the fault detection of rolling bearing.

No abstract

This study suggests a simple, quick and non-destructive method for investigation of mechanical parameters (dynamic Young's modulus, dynamic shear modulus and Poisson's ratio) detection for rectangular plate structures in laminated composites which only utilizes the fundamental resonant frequency in flexural and torsional modes, mass and dimensions of structures. The method is based on the impulse excitation technique (IET) to pick up the fundamental resonant frequency and then the corresponding formulas are applied to evaluate the mechanical parameters. Numerical simulations using finite element method (FEM) and experimental investigations of several cases based on IET are introduced to verify the accuracy of the IET formulas. The results show that the IET is applicable for mechanical parameters identification for laminated composites plates. The method is expected to detect mechanical parameters of other more complicated structures.

This paper focuses on the fault diagnosis for NC machine tools and puts forward a fault diagnosis method based on kernel principal component analysis (KPCA) andk-nearest neighbor (kNN). A data-dependent KPCA based on covariance matrix of sample data is designed to overcome the subjectivity in parameter selection of kernel function and is used to transform original high-dimensional data into low-dimensional manifold feature space with the intrinsic dimensionality. ThekNN method is modified to adapt the fault diagnosis of tools that can determine thresholds of multifault classes and is applied to detect potential faults. An experimental analysis in NC milling machine tools is developed; the testing result shows that the proposed method is outperforming compared to the other two methods in tool fault diagnosis.

Impulse excitation technique is a simple, convenient, and standard nondestructive method to detect mechanical parameters (dynamic Young’s modulus and dynamic torsional modulus) that only utilizes the first-order flexural resonant frequency, first-order torsional resonant frequency, and dimensions of structures. However, the mechanical parameter detection formulas are well established only for standard uniform specimens with uniform rectangular and circular cross sections. This study suggests a simulation-based method to detect mechanical parameters. A response surface method (RSM) is introduced to design numerical simulation experiments to build up experimental formulas to detect mechanical parameters. Numerical simulations are performed by the finite element method (FEM) to obtain enough simulation data for RSM analysis. After calculations, the two relationships (experimental formulas) can finally be obtained, i.e., the relationship of the dynamic Young’s modulus and first-order flexural resonant frequency with respect to dimensions of structures and the relationship of the dynamic torsional modulus and first-order torsional resonant frequency associated with dimensions of structures. Numerical simulations and experimental investigations show that the simulation-based method can be used to detect mechanical parameters in I-beams and hollow cylinders. More generally, this method can be further developed to detect the mechanical parameters of more complex structures than standard uniform specimens using a combination of FEM simulation and RSM.