Analisis data rawak terikan-lesu menggunakan kaedah statistik purata-tergerak

Abstract

This paper presents the analysis of fatigue road loading which was measured on a component of an automobile suspension system. Thus, this significant technique was introduced for preserving data associated to the underlying probabilistic properties that are related to the fatigue damage. The analysis was based on the fatigue damage potential, which was related to the random variables in a time series data. Using this data type, the analysis was performed by means of the statistical method, such as the Moving Average technique. This model can then be applied to estimate the data trend by using the related Moving Average models in order to reduce a random variation of fatigue data. In this study, the data were experimentally measured on an automobile suspension system which was travelling over a public road surface. During the measurement, the data collection was performed at the sampling rate of 200Hz in the 300 second of the record length. For the analysis, the Moving Average method was applied to the fatigue data in order to determine the parameters related to Mean Absolute Percentage Error (MAPE), Mean Absolute Deviation (MAD) and Mean Squared Deviation (MSD). In addition, the significant analysis of data correlation based on Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) was also discussed. Finally, it is suggested that this method provided a good platform to process the changeable random fatigue strain loading in order to produce a stationary data, with the removal of the nonstationary parts.