基于数据关联规则与动态称重传感器布局.docVIP

基于数据关联规则的动态称重传感器布局* 王 翥 侯甲童 (哈尔滨工业大学(威海)信息科学与工程学院, 威海 264209) 摘 要: 动态称重是汽车在高速路行驶中获取汽车真实轴重的有效方法。本文介绍了压电传感器的原理以及简化的动态称重模型。在使用压电传感器的基础上, 应用时序步进法对多个压电传感器进行布置, 并对布置进行局部调整, 来达到较为合理的目标。应用数据关联规则的方法, 选取两个较好的布置来进行仿真实验, 用Gass-newton法来求动态轴重信号模型, 得到更接近真实轴重的值。在不同初始相位下, 对一定速度下不同布局的误差进行仿真, 并对结果进行比较。结果显示精度均在一定的可以接受的范围之内, 并且融合程度与多个传感器的布置优劣程度紧密相关。 关键词: 动态称重；压电传感器；数据关联规则；Gass-newton 中图分类号: TP202　　　文献标识码: A　　　国家标准学科分类代码: 510.8060 Layout of weigh-in-motion sensor based on data association rules Wang Zhu Hou Jiatong (School of Information Science and Engineering, Harbin Institute of Technology (Weihai), Weihai 264209, China) Abstract: Weigh-in-motion (WIM) is a better way that can gain the true weight of vehicle which is moving on the highway at high speed. In this paper, the principle of piezoelectric sensor is introduced, and the WIM signal model is simplified. Timing stepping based on piezoelectric sensors is used to distribute the multi-sensor and adjust the distribution partly, and can get more reasonable goal. The application of data association rules is used to get two better distributions for simulation, and then the WIM model and the expected value which is closer to the true axle load can be gained by the Gauss-Newton method. The simulation is carried out under the different speeds errors, when the initial phases are different according to different layouts, and the result is compared. The result shows that the accuracy is within the receivable range, and the correlation function is related to the distribution’s degree of advantages and disadvantages of multi-sensor closely. Keywords: WIM; piezoelectric sensor; data association rule; Gauss-Newton method 1 引 言 目前对车辆的载重监测有静态称重和动态称重两种方法。静态称重虽然精度高, 但其规模较大, 引人注目, 往往使许多超载车辆易于逃避, 同时静态称重还容易造成交通不畅, 难以发挥高速公路的优越性, 因此常采用动态称重[1]。 车辆动态称重的主要方式有两种:整车计量和轴计量。应用整车计量方式, 需要比较大的秤台, 这大大增加了工程造价和难度, 所以这种方式使用得越来越少。目前较为流行的动态汽车称重系统是轴重称量, 既分别测出车辆的各个轴的轴重, 再由测试系统计算出整车重量。 当汽车以一定的速度通过称重台面时, 不仅轮胎对台面的作用时间很短, 而且作用在台面上的力除真实轴重外, 还有许多因素产生的干扰力, 如车速、车辆自身谐振、路面激励