中文摘要：

为了提高轮胎泵浦噪声的预测精度，在Hayden模型的理论基础上，考虑轮胎横向沟槽的位置效应，提出了一种改进的轮胎泵浦噪声预测模型——非等强度多源模型。采用激光双三角测量技术解决了预测模型中沟槽体积变化参数难以获取这一难点，通过激光双三角测量系统测量了具有纵、横沟槽胎面花纹的轮胎在静态受压时的沟槽轮廓信息，并模拟计算了80km·h^-1速度下沟槽体积变化与时间的关系曲线。考虑到静态测量结果与实际动态结果之间的差异，根据相关有限元计算研究成果对静态测量结果进行了修正，并将修正后的沟槽体积变化数据分别代入非等强度多源模型和Hayden模型进行轮胎噪声的预测。为了验证提出模型的有效性，采用实验室转鼓法对轮胎进行了噪声测量，并将测量结果与2种模型的预测结果进行对比。研究结果表明：激光双三角测量系统能准确测量轮胎花纹沟槽的轮廓变形，经过计算以及修正后的沟槽体积变化量与已有研究成果一致；Hayden模型的预测结果比实验室转鼓法测量结果高约10dB，且只能预测总声压级，无法对频谱结果进行预测；非等强度多源模型预测的频谱曲线与实测频谱曲线基本一致，噪声总声压级以及频谱各阶峰值的预测误差均在±3dB以内，具有较高的预测精度。

英文摘要：

To improve the prediction precision of tire pumping noise, an optimized prediction model named non-equal intensity multiple sound sources model based on theory of Hayden model was proposed. Position effect of tire lateral grooves was considered by this model. The key problem about how to get the parameters of tire grooves volume change in the prediction model was solved by laser dual-triangulation measurement technology. The grooves profile data of a tire with lateral and longitudinal grooves under static loading were measured with laser dual- triangulation measurement system and the relationship between grooves volume change and time at 80 km~ h 1 was simulated and calculated. Considering the difference between the static measurement result and the actual result under dynamic condition, a correction on static measurement result was made according to the FEM results. The tire noise was predicted by substituting the data of corrected grooves volume change to optimized model and Hayden model respectively. In order to validate the optimized model, tire noise was tested in a drum laboratory,and the comparison was made among the test results and prediction results of two models. The results show that the deformation of grooves profile can be measured by laser dual-triangulation measurement system accurately, and grooves volume change amount after calculation and correction is in accordance with the existing research results. The prediction result of Hayden model is higher than test result by about 10 dB. Hayden model can only make a prediction of total sound pressure level, and it cannot predict the spectrum results. The spectrum curves predicted by non-equal intensity multiple sound sources model are in accordance with test curves, and the errors of total sound pressure level and peaks at each order are within ±3 dB, which shows a higher prediction precision.