基于虚拟简支梁法的桥梁动态称重研究

Vehicles' BWIM based on virtual simply-supported beam method

Moses算法是桥梁动态称重(BWIM)技术中最可靠的算法之一,是目前各商业BWIM系统的基础。然而,受影响线标定的约束,当前的商用BWIM系统仅适用于短跨径桥梁。针对这一情况,提出了虚拟简支梁法。该方法利用桥梁上的某一区段的隔离应变计算车辆的轴重和总重,不受桥梁跨径的限制。建立了跨径为20 m和40 m的简支T梁桥有限元模型,基于车桥耦合振动理论模拟获得桥梁响应,并利用提出的新方法识别了车辆的轴重和总重。分析了路面平整度、车辆行驶速度等因素对识别精度的影响。结果表明:对于较短跨径的20 m T梁桥,常见三轴车的总重识别误差平均值在2%左右,五轴车的总重识别误差平均值低于1%,精度都稍优于传统Moses算法;而对于传统Moses算法不适用的40 m T梁桥,该方法识别车辆总重的误差平均值仍可控制在3%以内,表明该方法不受桥梁跨径的限制,具有更广的应用前景。

Moses algorithm is one of the most reliable algorithms in bridge weigh-in-motion（ BWIM） techniques,and is the basis of commercial BWIM systems. However,restricted by calibration of influence lines, the current commercial BWIM systems are only applicable to short-span bridges. Here,aiming at this situation,the virtual simplysupported beam method was proposed. With this method,the strain of an isolated section of a bridge was used to calculate the axle load and the weight of a vehicle not restricted by bridge span. The finite element models of two simply-supported T-beam bridges with spans of 20 m and 40 m were built,respectively. Based on the theory of vehicle-bridge coupled vibration,the two bridges＇ responses when a vehicle passes through them were obtained. Then the proposed method was used to identify the axle load and the weight of a vehicle. The influences of road surface roughness and vehicle speed,etc. on the identified accuracy were analyzed. The results showed that for the T-beam bridge with a shorter span of 20 m,the average identification errors for the weight of a three-axle vehicle and a five-axle one are about 2% and less than 1%,respectively,the accuracies are slightly higher than those using the traditional Moses algorithm; for the T-beam bridge with a longer span of 40 m to which the traditional Moses algorithm is not applicable,the average identification errors for vehicles＇ weights can be controlled within 3% by using the proposed method; so,the proposed method is not restricted by bridge span and has wider application prospects.