4.2. Kurtosis Evaluation
To investigate whether the random vibration signals were
Gaussian, a kurtosis parameter of the z-axis frequencyweighted
floor accelerations were evaluated for different road
surfaces at 20 km/h and 80 km/h.
Figure 7a shows the variations of kurtosis and VDV with
changes in IRI (from Table 4) at 20 km/h. The right axis of
each graph corresponds to VDV. It can be seen that kurtosis
values increase as the IRI increases. This indicates a deviation
of the acceleration signals from the Gaussian distribution as the
IRI increases. On all the roads, VDV values increased as the
road roughness increased. As expected, driving on rough road
surfaces induces higher peaks and impulses. This resulted in
more kurtosis and VDV values and less objective driver comfort.
Hence, road roughness could be compensated through
slowing down and thereby improving the ride quality. Similar
results may be concluded from Fig. 7b, which shows variations
of kurtosis and VDV versus road roughness at 80 km/h.
4.3. SEAT Values Evaluation
Figure 8 shows the comparison of V DVseat and V DVbase for
driving over road surfaces at specified speeds. Data points lie
under a 45-degree diagonal starting at the origin. It shows
SEAT values of less than 100% and isolation of vibrations.
Figure 8. Comparison of the V DVseat and V DVbase values on road surfaces.
4.4. Frequency Analysis of Vibration
Signals
All signals (except on the suburban road) were acquired over a
period of 60 s, and the frequency span of analysis was 100 Hz.
The PULSE analyzer was adjusted in a way that an arbitrary
number of 3200 lines were implemented in Fast Fourier Transform
(FFT) analysis to achieve a high-frequency resolution of
31.25 mHz (100/3200).
The analyzer automatically detected the mean square of each
signal and divided it by the bandwidth to calculate the PSD
value. Such narrowband analysis shows high coherency, close
to unity, between seat-surface and seat-base signals. The Frequency
Response Function (FRF) analysis between these signals
for the pavement road, at a speed of 20 km/h, is presented
in Fig. 9. This graph implies that the seat structure was a good
isolator of vibration below 30 Hz, while after that, the signal
was amplified, but it was not a critical issue because, as shown
4.2. Kurtosis Evaluation
To investigate whether the random vibration signals were
Gaussian, a kurtosis parameter of the z-axis frequencyweighted
floor accelerations were evaluated for different road
surfaces at 20 km/h and 80 km/h.
Figure 7a shows the variations of kurtosis and VDV with
changes in IRI (from Table 4) at 20 km/h. The right axis of
each graph corresponds to VDV. It can be seen that kurtosis
values increase as the IRI increases. This indicates a deviation
of the acceleration signals from the Gaussian distribution as the
IRI increases. On all the roads, VDV values increased as the
road roughness increased. As expected, driving on rough road
surfaces induces higher peaks and impulses. This resulted in
more kurtosis and VDV values and less objective driver comfort.
Hence, road roughness could be compensated through
slowing down and thereby improving the ride quality. Similar
results may be concluded from Fig. 7b, which shows variations
of kurtosis and VDV versus road roughness at 80 km/h.
4.3. SEAT Values Evaluation
Figure 8 shows the comparison of V DVseat and V DVbase for
driving over road surfaces at specified speeds. Data points lie
under a 45-degree diagonal starting at the origin. It shows
SEAT values of less than 100% and isolation of vibrations.
Figure 8. Comparison of the V DVseat and V DVbase values on road surfaces.
4.4. Frequency Analysis of Vibration
Signals
All signals (except on the suburban road) were acquired over a
period of 60 s, and the frequency span of analysis was 100 Hz.
The PULSE analyzer was adjusted in a way that an arbitrary
number of 3200 lines were implemented in Fast Fourier Transform
(FFT) analysis to achieve a high-frequency resolution of
31.25 mHz (100/3200).
The analyzer automatically detected the mean square of each
signal and divided it by the bandwidth to calculate the PSD
value. Such narrowband analysis shows high coherency, close
to unity, between seat-surface and seat-base signals. The Frequency
Response Function (FRF) analysis between these signals
for the pavement road, at a speed of 20 km/h, is presented
in Fig. 9. This graph implies that the seat structure was a good
isolator of vibration below 30 Hz, while after that, the signal
was amplified, but it was not a critical issue because, as shown
การแปล กรุณารอสักครู่..