TY - BOOK
T1 - Modeling the pothole road
AU - Backx, M.A.F.
A2 - Nijmeijer, H.
A2 - Liebregts, R.M.J.
N1 - Internship report. - DC 2013.056
PY - 2013
Y1 - 2013
N2 - DAF manufactures and sells trucks in Brazil. The roads in Brazil correspond to a heavy duty load
spectrum. This is due to the large holes in the roads, which exert a heavy load on the truck. Quality,
durability and cost are important factors for designing DAF trucks. In the design process it is
important to simulate or test the loads which are exerted on a truck. In order to replicate the severe
conditions in Brazil the pothole road on the test track is used. This is a 200m long concrete road with
circular and elliptical holes of varying depths. The problem with this pothole track is, that there is
no road data available which describes it. With a model of the road, the response of the truck can be
simulated. By comparing the outcome of the simulation with measurement data of the test track and
of the roads of Brazil. The conditions in Brazil can be replicated by means of simulation and testing.
In order to model the pothole road, measurements are performed at the test track. Every hole in this
road is measured. These measurements are used to design two road descriptions in Matlab. The first
road description is method A. This is an altitude map and is based on a rectangular grid. The pothole
road is made by setting every grid element at a specific height. The tracks of the trucks tyres are interpolated
between the grid points. The resulting interpolated functions hold the information, a tyre
encounters while driving over the pothole road.
The second road description, method B. Is based on mathematically described circles and ellipsoids.
The trigonometric properties of the circles and ellipsoids are used to interpolate the tyre tracks on
the pothole road. The resulting interpolated functions hold the information, a tyre encounters while
driving over the pothole road.
Each road description serves its own purpose. Besides the different approach of designing the road
descriptions, they result in similar front axle simulation responses. In order to validate the pothole
road model, the simulation responses are compared with measurement data of a truck driving on the
pothole road. This validation is mainly done by comparing the PSDs of the front axle accelerations
in z-direction. In order to cope with variations in conditions, and variations in truck models, the vehicle
simulation model parameters are adjusted. The vertical tyre stiffness and front axle damping
are adjusted such that the PSD of the simulation data is similar to the PSD of the measurement data.
The tuning is done with a Response Surface Methodology, a key component of this methodology is
the Design Of Experiments. These statistical tools are used to minimize the difference between the
simulation and measurement data. The optimum model parameters are used to validate the road descriptions.
The PSD plots of different measurements are combined with simulations responses. This
shows that there is a lot of dispersion between the measurements. The simulation responses fits in
the dispersion region of the measurements. So the simulations on the pothole road result in front axle
accelerations that are similar to a truck driving on the pothole road.
AB - DAF manufactures and sells trucks in Brazil. The roads in Brazil correspond to a heavy duty load
spectrum. This is due to the large holes in the roads, which exert a heavy load on the truck. Quality,
durability and cost are important factors for designing DAF trucks. In the design process it is
important to simulate or test the loads which are exerted on a truck. In order to replicate the severe
conditions in Brazil the pothole road on the test track is used. This is a 200m long concrete road with
circular and elliptical holes of varying depths. The problem with this pothole track is, that there is
no road data available which describes it. With a model of the road, the response of the truck can be
simulated. By comparing the outcome of the simulation with measurement data of the test track and
of the roads of Brazil. The conditions in Brazil can be replicated by means of simulation and testing.
In order to model the pothole road, measurements are performed at the test track. Every hole in this
road is measured. These measurements are used to design two road descriptions in Matlab. The first
road description is method A. This is an altitude map and is based on a rectangular grid. The pothole
road is made by setting every grid element at a specific height. The tracks of the trucks tyres are interpolated
between the grid points. The resulting interpolated functions hold the information, a tyre
encounters while driving over the pothole road.
The second road description, method B. Is based on mathematically described circles and ellipsoids.
The trigonometric properties of the circles and ellipsoids are used to interpolate the tyre tracks on
the pothole road. The resulting interpolated functions hold the information, a tyre encounters while
driving over the pothole road.
Each road description serves its own purpose. Besides the different approach of designing the road
descriptions, they result in similar front axle simulation responses. In order to validate the pothole
road model, the simulation responses are compared with measurement data of a truck driving on the
pothole road. This validation is mainly done by comparing the PSDs of the front axle accelerations
in z-direction. In order to cope with variations in conditions, and variations in truck models, the vehicle
simulation model parameters are adjusted. The vertical tyre stiffness and front axle damping
are adjusted such that the PSD of the simulation data is similar to the PSD of the measurement data.
The tuning is done with a Response Surface Methodology, a key component of this methodology is
the Design Of Experiments. These statistical tools are used to minimize the difference between the
simulation and measurement data. The optimum model parameters are used to validate the road descriptions.
The PSD plots of different measurements are combined with simulations responses. This
shows that there is a lot of dispersion between the measurements. The simulation responses fits in
the dispersion region of the measurements. So the simulations on the pothole road result in front axle
accelerations that are similar to a truck driving on the pothole road.
M3 - Report
T3 - D&C
BT - Modeling the pothole road
PB - Eindhoven University of Technology
CY - Eindhoven
ER -