Abstract:

Tire rolling resistance is an important factor affecting vehicle fuel economy, which is mainly due to the energy loss caused by the hysteresis effect of rubber materials. For passenger cars using radial tires, about 3.4% to 6.6% of fuel consumption is used to overcome tire rolling resistance, so the topic of reducing vehicle fuel consumption by reducing tire rolling resistance has received more and more attention from scholars. The purpose of this paper is to establish a more accurate and effective tire rolling resistance prediction model by using acceleration signal combined with intelligent tire technology.

In this paper, the 205/55/R16 passenger car radial tire is taken as the research object. Firstly, based on the contribution rate of rolling resistance, the structure of the radial tire is simplified reasonably, and the finite element model of the tire is established by ABAQUS finite element simulation software and material parameterization method. The UAMP subroutine tire is used to control the angular velocity of the tire to obtain the steady-state free rolling angular velocity of the tire and extract the rolling resistance data. Through finite element analysis and control variable method, the rolling resistance of tire finite element model under variable load, vehicle speed and tire pressure is studied. The validity of the finite element model is verified by the stress and strain characteristics of tire joints.

Secondly, the acceleration data of the nodes at the central axis of the tire lining under various compound working conditions are extracted, and the acceleration of the nodes is converted from the body coordinate system to the acceleration body coordinate system using the coordinate transformation matrix, and the longitudinal, lateral and vertical acceleration curves are obtained. Comparing the response degree of different signals to rolling resistance, the longitudinal and vertical acceleration signals are selected as the observation signals. The generation mechanism of tire rolling resistance is analyzed. Yule-Walker frequency domain analysis method is used to calculate the power spectral density of acceleration signals. The relationship between signal power and frequency is estimated through the correlation of signals. Combined with tire pressure, vehicle speed and load, a tire rolling resistance estimation model based on partial least squares regression algorithm is built.

Finally, the fitting effect of the model can be approximated according to the estimation results of the model under 20 test conditions of variable load, vehicle speed and tire pressure. The mean square error of the tire rolling resistance estimation algorithm based on acceleration signal is 0.318 3, and the goodnessof fit is 0.967 6. Under the same data set, the mean square error and goodness of fit of the tire rolling resistance estimation algorithm, which only uses vehicle speed, tire pressure and load as input variables, are 0.352 4 and 0.941 9. The results show that compared with the traditional physical model of rolling resistance, the fitting effect of the regression equation combining the tire acceleration signal and driving parameters is better than that of using only the driving parameters, and the prediction result is more accurate, which may provide some references for the research of rolling resistance.