Premium
On the Statistical Distribution of Road Vehicle Vibrations
Author(s) -
GarciaRomeuMartinez ManuelAlfredo,
Rouillard V.
Publication year - 2011
Publication title -
packaging technology and science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.365
H-Index - 50
eISSN - 1099-1522
pISSN - 0894-3214
DOI - 10.1002/pts.950
Subject(s) - kurtosis , weibull distribution , skewness , standard deviation , vibration , normal distribution , gaussian , root mean square , random vibration , range (aeronautics) , mathematics , mean squared error , statistics , engineering , physics , quantum mechanics , aerospace engineering , electrical engineering
SUMMARY This paper presents an alternative method for characterizing the random vibrations produced by transport vehicles. The paper discusses the significance and limitations of the average power spectral density and explains why it is not always adequate as the sole descriptor of road vehicle vibrations as the process generally tends to be non‐stationary and non‐Gaussian. The paper adopts an alternative analysis method, based on the statistical distribution of the moving root‐mean‐square (rms) vibrations, as a supplementary indicator of overall ride quality. A variety of sample vibration records, collected from various vehicle types and routes in Spain and Australia, were used to investigate the suitability of various mathematical models, based on the Weibull distribution. It shows that the model can also effectively describe the statistical parameters of the process, namely the mean, median, standard deviation, skewness and kurtosis. The paper proposes a single mathematical model that can accurately describe the statistical character of the non‐stationary random vibrations generated by road vehicles in general. The proposed generic distribution model, based on the Weibull distribution, was developed to afford additional control over various aspects of the shape of the distribution function. The model was found to be general enough to be able to produce a range of well‐known distributions. Curve‐fitting results using the sum‐of‐squared error (least squares) optimization were found to produce non‐convergent results, which required inclusion of the mean, median, standard deviation, skewness and kurtosis in the optimization algorithm. The paper also shows how the model is capable of accurately describing the statistical parameters of the process, namely the mean, median, standard deviation, skewness and kurtosis. This result is relevant not only for the characterization of ride quality but also for the accurate synthesis of road vehicle vibrations in the laboratory. The results can be used to assist in developing a novel method for simulating non‐stationary (modulated) vibration in the laboratory. The rms distribution function can be used to create an rms level schedule that will enable the synthesis of random vibrations with varying rms levels to better represent the road transport vibration process. Copyright © 2011 John Wiley & Sons, Ltd.