Research on on-board dynamic weighing algorithm based on two-degree-of-freedom 1/4 vehicle model

作     者：Fu, Hongxun Zhang, Yu Gang, Xianyue Qiao, Huanbo Wang, Yan Ku, Laiyun 

作者机构：Shandong Univ Technol Sch Transportat & Vehicle Engn 12 Zhangzhou Rd Zhangdian 255049 Zibo Peoples R China 

出 版 物：《INTERNATIONAL JOURNAL OF HEAVY VEHICLE SYSTEMS》 (国际重型机动车系统杂志)

年 卷 期：2021年第28卷第6期

页      面：792-807页

核心收录：

学科分类：08[工学] 0802[工学-机械工程] 0823[工学-交通运输工程] 

主　　题：on-board weighing system two-degree-of-freedom 1 4 vehicle model dynamic weighing algorithm wavelet threshold denoising algorithm BP neural network 

摘      要：The moving vehicle will be disturbed in many aspects, resulting in the dynamic weighing accuracy of the airborne weighing system being significantly lower than the static accuracy. In order to improve the dynamic weighing accuracy of the system, this paper designs a dynamic weighing algorithm based on wavelet threshold denoising and BP neural network. Firstly, a two-degree-of-freedom 1/4 vehicle model was built to obtain the vehicle dynamic distance data. Then, the wavelet threshold denoising algorithm was used to denoise the dynamic distance data. Finally, the BP neural network was constructed with the signal of vehicle speed, acceleration signal and denoised weight signal as the input layer to reduce the impact of the speed and acceleration on the weight signal. The results show that after the processing of dynamic weighing algorithm, the dynamic weighing error of vehicle is less than 2%, and the algorithm meets the accuracy requirements, and has high universality.