時間分解能に優れる周波数解析手法を用いた車体振動の入力輪推定手法

抄録

Evaluation for ride comfort in vehicle development is mainly performed by a testing expert who has been trained. Hence determining of which wheel receives road input is also determined by him when comfortable or uncomfortable vehicle vibration happens. If we can understand the relationship between input and output of the vehicle vibration quantitatively instead of deeming with human sensation, it could be one of ways that we understand human sensation for ride comfort more and it may lead to a development of new components and control methods which affect human sensation directly. Frequency analysis like Fourier transform, Short-time Fourier transform and Wavelet transform with magnitude-squared coherence are often used for finding the relationship between input and output signal. However those frequency analysis methods and magnitude-squared coherence aren’t suitable for non-stationary vibration like which is measured when a vehicle runs on rough road and off road. In this report the method to find the relationship between input from road and non-stationary output at a vehicle body is described with Hilbert-Huang Transform (HHT) which suits non-stationary and non-liner signals and ideas which comes from magnitude-squared coherence.