面向复杂场景的车辆动态称重方法  

Dynamic Vehicle Weighing Methods for Complex Scenarios

作　　者：韩金波 周鹏飞 杨晓林 钟良骥 陈秋童 魏喆 廖海斌 HAN Jin-bo;ZHOU Peng-fei;YANG Xiao-lin;ZHONG Liang-ji;CHEN Qiu-tong;WEI Zhe;LIAO Hai-bin(Wuhan Taiwoz Information Technology Co.,Ltd.,Wuhan 430058 China;Wuhan Dongwohuida Technology Co.,Ltd.,Wuhan 430058,China;School of Computer Science and Technology,Hubei University of Science and Technology,Xianning 437100,China;School of Electronic and Electrical Engineering,Wuhan Textile University,Wuhan 430200,China)

机构地区：[1]武汉泰沃滋信息技术有限公司,湖北武汉430058 [2]武汉东沃慧达科技有限公司,湖北武汉430058 [3]湖北科技学院计算机科学与技术学院,湖北咸宁430070 [4]武汉纺织大学电子与电气工程学院,湖北武汉430200

出　　处：《湖北科技学院学报》2024年第6期140-146,共7页Journal of Hubei University of Science and Technology

基　　金：湖北省自然科学基金(2021CFB388);大学生创新创业计划项目(2013017)。

摘　　要：在动态称重系统的实际应用中,车辆重量的测量结果往往受到多种外部环境因素的显著影响,如车型差异、胎压变化和车速波动等,这些因素导致测量结果的波动较大,影响了称重的准确性。为了应对这一挑战,本文提出了一种创新的基于因子分析模型的车辆动态称重方法,旨在有效地分离和减少这些干扰因素对称重结果的影响。首先,对采集到的特征矩阵运用双线性模型进行深入的因子分析。通过这种方法,能够有效地分离出车辆外部因子,从而提取出与车辆本身重量更为紧密相关的、近似车辆无关的车重特征。随后,利用这些经过因子分离处理后的车重信号进行车辆重量的估算。通过此方法,能够更准确地反映车辆的真实重量,同时减少了由于外部环境因素引起的误差。实验结果表明,本文所提出的方法在动态称重中表现出色。它不仅显著改善了称重的稳定性,而且成功地将称重误差控制在了5%以下的较低水平,这充分验证了该方法的有效性和实用性。In the practical application of dynamic weighing system,the measurement results of vehicle weight are often significantly affected by a variety of external environmental factors,such as model differences,tire pressure changes and vehicle speed fluctuations,etc.,which lead to large fluctuations in measurement results and affect the accuracy of weighing.In order to address this challenge,this paper proposes an innovative dynamic vehicle weighing method based on factor analysis model,which aims to effectively separate and reduce the influence of these interfering factors on weighing results.Firstly,a bilinear model was used to conduct in-depth factor analysis on the collected feature matrix.Through this method,the external factors of the vehicle can be effectively separated,so as to extract the vehicle weight characteristics that are more closely related to the weight of the vehicle itself and have nothing to do with the approximate vehicle.Subsequently,the vehicle weight signals after factor separation were used to estimate the vehicle weight.With this method,the true weight of the vehicle can be more accurately reflected,while the error caused by external environmental factors is reduced.The experimental results show that the proposed method performs well in dynamic weighing.It not only significantly improves the stability of weighing,but also successfully controls the weighing error at a low level of less than 5%,which fully verifies the effectiveness and practicability of the method.

关 键 词：动态称重 因子分析 双线性模型 EM算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]