Decomposition of the acceleration levels distribution of a road transport into a sum of weighted Gaussians: Application to vibration tests. Issue 8 (10th May 2018)

• Record Type: Journal Article
• Authors: Bonnin, Anne‐Sophie; Nolot, Jean‐Baptise; Huart, Victor; Pellot, Jérôme; Krajka, Nicolas; Odof, Serge; Erre, Damien
• Is Part Of: Packaging technology and science. Volume 31:Issue 8(2018)
• Page Start: 511
• Page End: 522
• Publication Date: 2018-05-10
• Subjects: Packaging -- Periodicals; 688.8
• DOI: 10.1002/pts.2375
• Languages: English
• ISSNs: 0894-3214

Abstract

Abstract : The usual method to simulate vertical vibrations generated by road transport in laboratory is to use the average power spectral density. With this method, the distribution of the acceleration levels throughout the test is a Gaussian which does not conform to the reality of a transport. This study proposes an improvement of the classical method; this improvement permits to simulate the power spectral density and the distribution of acceleration levels by using a conventional device. The main idea of this method is to decompose the distribution of an actual road transport by using a sum of weighted Gaussians. Then, we apply the power spectral density with the root mean square acceleration (grms ) level and duration corresponding to each Gaussian calculated from the weighted sum. We show that the weighting coefficients correspond to the time fraction of the test. Over the total duration of the test, we then retrieve the acceleration levels distribution of the actual transport. This new test method is experimentally validated with 2 examples of actual transports. Abstract : This paper introduces a method to generate the same probability density function of acceleration levels occurrence in laboratory as in an actual road transport by using its own power spectral density. The study is based on an iterative method that permits to determine the durations and levels of application of the actual power spectral density corresponding to each weighted Gaussian.