The article presents a novel non-linear multivariate and multiscale statistical process monitoring and signal denoising method which combines the strengths of the Kernel Principal Component Analysis (KPCA) non-linear multivariate monitoring approach with the benefits of Ensemble Empirical Mode Decomposition (EEMD) to handle multiscale system dynamics. The proposed method which enables us to cope with complex even severe non-linear systems with a wide dynamic range was named the EEMD-based multiscale KPCA (EEMD-MSKPCA). The method is quite general in nature and could be used in different areas for various tasks even without any really deep understanding of the nature of the system under consideration. Its efficiency was first demonstrated by an illustrative example, after which the applicability for the task of bearing fault detection, diagnosis and signal denosing was tested on simulated as well as actual vibration and acoustic emission (AE) signals measured on purpose-built large-size low-speed bearing test stand. The positive results obtained indicate that the proposed EEMD-MSKPCA method provides a promising tool for tackling non-linear multiscale data which present a convolved picture of many events occupying different regions in the time-frequency plane.
COBISS.SI-ID: 11953179
A new, fast and flexible, time-dependent, one-dimensional numerical model was developed in order to study in detail the operation of an active magnetic regenerator (AMR). The model is based on a coupled system of equations (for the magnetocaloric material and the heat-transfer fluid) that have been solved simultaneously with the software package MATLAB. The model can be employed to analyze a wide range of different operating conditions (mass-flow rate, operating frequency, magnetic field change), different AMR geometries, different magnetocaloric materials and heat-transfer fluids, layered and single-bed AMRs, etc. This paper also presents an optimization of the AMRA news geometry, where the AMR consists of a packed-bed of grains (spheres) of gadolinium (Gd). The optimization of the mass-flow rate and the operating frequency of the AMR were performed by studying five different diameters of Gd spheres.
COBISS.SI-ID: 11935003
Although substantial progress has been made in the field of occupant safety in recent years, protection of children is still not optimal. With the development of computer technology in the last decade, FEM (finite element method) has become a powerful virtual tool for simulating dynamic response of the human body under various conditions. This study presents the development of a detailed FE model of a 3-year old child (3YO) for the analysis of neck injuries caused by vehicle accidents. Due to the lack of 3D geometrical skeleton model data for a 3YO child, a new FE model was developed, as a scaled version of the previously developed adult FE model, based on the anthropometric data for a 3YO child. The detailed model contains neck anatomical structures which influence the dynamic response that are likely to be injured during a vehicle impact (cervical vertebrae, intervertebral discs, ligaments, muscles). Soft tissue material properties that are predominantly determined for adult persons were obtained from the literature. Later, further optimization of these material properties was performed in order to gain better agreement between experimental and simulation results. Compared to other FE models, the anatomical connection between the neck and the torso is left uninterrupted, thus allowing a more precise analysis of the impact response for all impact directions. The presented model has been validated by means of published crash test results with child dummies. The response of the FE model correlates reasonably well, which confirms validity of this model.
COBISS.SI-ID: 12118043