English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Engineering surfaces are comprised of different wavelength bands commonly referred to as roughness, waviness and form. The Gaussian filter is a common technique to separate surface roughness profile but it has some drawbacks. This paper presents an approach of utilizing Empirical Mode Decomposition (EMD) to separate surface roughness profile from raw contour. The EMD describes a time-signal with a series of Intrinsic Mode Functions (IMFs) and the residue. The residue and IMFs of frequency less than 10 within an evaluation length, are used for reconstructing EMD reference line. To achieve more accurate mean line, an Amplitude Weighting Factor (AWF) μ is applied to the IMF whose frequency is maximum among the selected IMFs. The EMD reference line is thus modified as EMD mean line. Surface roughness profile can be further separated and roughness parameters can be calculated. The comparison of EMD method and Gaussian filter is also presented. Introduction A typical engineering surface consists of roughness, waviness and form [1], which influences the performance of mechanical system [2]. To evaluate various parameters of surface profile, it is necessary to extract surface roughness profile. The separation of roughness profile is usually based on a mean line and filtering methods are commonly adopted to determine the mean line. The filters include 2RC filter, Gaussian filter, Spline filter and Wavelet filter [3]. Although the 2RC filter is still applied in traditional instruments, the filter has been deleted from ISO standard since its inherent defects. Gaussian filter is the most common filter in the assessment of surface morphology, but it suffers from edge distortion effects [4] [5]. The Spline and the Wavelet filters are recent advances in filtering of surface morphology. The Spline filter gets the reference line by a simple closed formula [6]. The Wavelet filter is a multi-resolution decomposition method but the analyzing results to a same engineering surface are different with different wavelet bases [7]. Empirical Mode Decomposition (EMD) is proposed by Huang and can be used to analyze nonlinear and time-varying signal [8]. The EMD is self-adaptive and doesn’t need any basis function. A complex signal can be decomposed into a series of Intrinsic Mode Functions (IMFs) and the residue. The decomposition is based on the local characteristic time scale of the signal. In theory, the EMD method can be used for decomposition of any type of time-signal, especially for processing non-stationary and nonlinear signals. So, the method was applied widely as soon as it was put forward. As shown in Figure 1 (a), a profilometer is used to measure a surface of a workpiece. The measured raw contour is shown in Figure 1(b) where the roughness profile rides the trend profile of raw contour. In this paper, the EMD method is used to decompose the raw contour. The decomposed signals with low frequency are selected to reconstruct the EMD mean line. The roughness profile is thus separated and roughness parameters are calculated. International Conference on Advances in Mechanical Engineering and Industrial Informatics (AMEII 2015) © 2015. The authors Published by Atlantis Press 406 (a) (b) Fig. 1. A profilometer (a), and the measured raw contour (b) Empirical Mode Decomposition of Raw Contour The EMD considers that a signal is composed of many components called IMFs. The implementation of EMD is a sifting process resulting in IMFs. For a signal x(t), the sifting process is described as follows [8]: 1) Identify all extrema of x(t). 2) Find the maximum envelope emax(t) of x(t) by passing a natural cubic spline through the maxima. Similarly find the minimum envelope emin(t) with the minima. 3) Compute the mean, r(t)=(emax(t)+ emin(t))/2. 4) Compute the new sequence mi(t)=x(t)−r(t). 5) Check whether mi(t) is an IMF. If mi(t) is not an IMF, repeat the loop on mi(t). If mi(t) is an IMF then set ci(t)=mi(t). The process will be terminated until either the residue, rn, becomes a monotonic function from which no more IMF can be extracted, or when the IMF, cn(t), or the residue, rn, becomes so small that it is less than the predetermined value [8]. Once one of these criteria is satisfied, n-empirical modes of the signal are extracted. Then we obtain an expression of x(t) as ) ( ) ( ) ( 1 t r t c t x n n

Separation of Surface Roughness Profile from Raw Contour based on Empirical Mode Decomposition