【作者】 张斌；

【导师】 赵华； 邓继平；

【作者基本信息】 湖南大学 ， 桥梁与隧道工程（专业学位）， 2022， 硕士

【摘要】 桥梁动态称重(Bridge Weigh-in-motion,BWIM)系统是一种基于桥梁实测响应进而反算过桥车辆轴重及总重量的技术,是一种具有重大潜力的交通荷载监测手段。传统的BWIM系统大多依赖于车轴探测器来确定过桥车辆的时空位置,并且称重传感器大多数均为接触式传感器。针对其存在的车轴探测器仅适用部分桥型和接触式传感器安装困难等不足之处,提出了一种基于改进虚拟轴算法和数字图像相关技术的无接触式桥梁动态称重系统。此外,当BWIM系统应用于正交异性钢桥面时,其车轮横向位置会严重影响轴重识别精度。基于此,提出了一种适用于正交异性钢桥面的车轮横向位置识别方法。本文主要从以下几个方面展开研究:(1)提出了一种基于正则化技术和速度迭代模型的改进虚拟轴算法,该算法完全摒弃了桥梁动态称重系统对车轴探测器的需求。在改进虚拟轴算法中,将车辆的虚拟轴和速度均作为算法的迭代变量,引入正则化技术并基于最小二乘法建立桥梁实测响应和理论响应的误差方程,使用最小轴重约束和最小轴距约束筛选出真实车轴可能存在的位置。最后,遍历求解不同假定速度下的误差函数以识别过桥车辆信息(车速、轴数、轴距、轴重及总重)。随后,基于实桥试验(标定试验和随机车流试验)验证了该方法的有效性,并探究了车辆速度、虚拟轴间距和正则化参数对算法识别精度的影响。实桥试验结果表明:改进虚拟轴算法可以提供与使用车轴探测器的Moses算法相接近的车辆信息识别精度。(2)提出了一种基于改进虚拟轴算法和数字图像相关技术(Digital Image Correlation,DIC)的无接触式桥梁动态称重系统。系统地推导了主流DIC技术的像素搜索算法,并验证了DIC技术室内/外位移测试精度。利用DIC技术测量得到的桥梁位移响应,基于改进虚拟轴算法识别过桥车辆信息,并通过缩尺模型试验验证了该方法的有效性。该方法在识别过桥车辆信息的过程中,仅使用到数字图像采集设备,实现了完全无接触式的桥梁动态称重系统。(3)提出了一种适用于正交异性钢桥面的车轮横向位置识别方法。该方法利用BWIM系统中称重传感器的响应信号来识别过桥车辆横向位置。通过建立正交异性钢桥面桥梁的有限元模型,提取其纵肋横向分布影响线,并基于横向分布影响线和最小二乘法,建立纵肋理论响应与实测响应的误差方程以识别车轮横向位置。随后,通过数值模拟和实桥试验验证了该方法的有效性。试验结果表明,提出的方法能有效识别过桥车辆的横向位置,且识别精度较高,适用范围广。更多还原

【Abstract】 Bridge weigh-in-motion(BWIM)system is a technology that calculates the axle weights and gross weight of vehicles crossing the bridge based on the measured response of the bridge.It is a traffic load monitoring method with great potential.Most traditional BWIM systems rely on axle detectors to determine the spatiotemporal position of vehicles crossing the bridge,and most of the load cells are contact sensors.Aiming at the shortcomings that the axle detector is only suitable for some bridge types and the installation of contact sensors is difficult,a contactless Bridge weigh-in-motion system based on extended virtual axle algorithm and digital image correlation technology is proposed.In addition,when the BWIM system is applied to an orthotropic steel bridge deck,the lateral position of its wheels will seriously affect the axle load identification accuracy.Based on this,a method for identifying the lateral position of wheels suitable for orthotropic steel bridge decks is proposed.This paper mainly focuses on the following aspects:(1)An extended virtual axle algorithm based on regularization technology and velocity iterative model is proposed,which completely removes the need for axle detectors in the BWIM.In the extended virtual axle algorithm,the virtual axle and speed of the vehicle are used as the iterative variables of the algorithm.The error equation of the measured response and the theoretical response of the bridge is established based on the least square method and the regularization technique.Then,the minimum axle weight constraint and the minimum wheelbase constraint are used to filter out where real axles might exist.Finally,the error function under different assumed speeds is traversed to identify the information of vehicles crossing the bridge(vehicle speed,number of axles,wheelbase,axle weight and gross weight).Subsequently,the effectiveness of the method is verified by field tests(calibration test and random traffic flow test),and the effects of vehicle speed,virtual axle spacing and regularization parameters on the recognition accuracy of the algorithm are explored.The real bridge test results show that the extended virtual axle algorithm can provide vehicle information recognition accuracy close to the Moses algorithm using axle detectors.(2)A contactless bridge weigh-in-motion system based on extended virtual axle algorithm and Digital Image Correlation(DIC)technology is proposed.The pixel search algorithm of mainstream DIC technology is systematically deduced,and the indoor/outdoor displacement test accuracy of DIC technology is verified.The bridge displacement response measured by DIC technology is used to identify the information of vehicles crossing the bridge based on the extended virtual axle algorithm,and the effectiveness of the method is verified by a scale model test.In the process of identifying the information of vehicles crossing the bridge,this method only uses digital image acquisition equipment,and realizes a completely contactless bridge weigh-in-motion system.(3)A method for identifying the lateral position of wheels suitable for orthotropic steel bridge decks is proposed.The method utilizes the response signal of the weigh station in the BWIM system to identify the lateral position of the vehicle crossing the bridge.By establishing the finite element model of the orthotropic steel deck bridge,the influence line of the longitudinal rib lateral distribution is extracted.Based on the lateral distribution influence line and the least square method,the error equation between the theoretical response and the measured response of the longitudinal rib is established to identify the lateral position of the wheel.Subsequently,the effectiveness of the method is verified by numerical simulation and field test.The test results show that the proposed method can effectively identify the lateral position of vehicles crossing the bridge,with high recognition accuracy and wide application range.更多还原