Some Approaches to the Rolling Wheels’ Dynamics Modelling in the Weight-in-Motion Problem

Some possibilities of fibre-optic sensors (FOS) application for measuring the weight of moving vehicles realized in weightin- motion (WIM) systems are discussed. As the first, the model of small-buried seismic sensor transient response excited by a car tyre interaction with asphalt-concrete road pavement is proposed. It is supposed that a seismic wave received by the sensor is the vertical component of surface Raleigh wave. The model is based on supposition that a tyre footprint is acceptable to consider as some array of point sources of these waves. The proper algorithms permit to vary different parameters of the array excitation, as to footprint dimensions, load distribution, car velocities and others. The set of Matlab codes is worked out for seismic pulses modelling and processing.

The second way considered is to simulate the FOS signal in the basis of differential equations describing a deformable wheel behaviour, or wheel oscillations, in order to identify relations with optoelectronic mechanical parameters. An attempt to find the mass of the vehicle is based on minimizing the discrepancy between the actual FOS signal and the solution of the differential equation. The accuracy of the evaluated weight depends on many external factors, the mathematical modelling of them are expressed in the numerical values of the coefficients and external stimuli. The influence of these factors are analysed and tested by simulations and field experiments.

One of ideas in dynamic weighing problem solution should consist in evaluation of position of virtual gravity centre of the vehicle in time. The processing algorithm of the data received from the FOS is proposed based on conception of database retaining in some reference system memory. Certain requirements concerning the elements and blocks of the algorithm are defined as well. The reference system is realized as the digital filter with the finite impulse response. The method to estimate the filter coefficients is worked out. Several experiments with this algorithm have been carried out for the vehicle identification with the reference loads adopted from real data.

The different factors have an influence on the measurement accuracy of FOS. The roadbed features, temperature, nonlinearities and delay effects in FOS are among them. The results of laboratory and field measurements with FOS responses to different axle’s loadings are presented. Charging and inertial characteristics of FOS under the impact of various external factors (protective cover, temperature, contact area, and installation mode especially) as well as their approximations are investigated. It is found that the final calibration of the FOS has to be done individually and only after it has been installed in the pavement. Certain methods and algorithms of linearization, as well temperature and dynamic errors compensation of FOS data are discussed.