关于车载动态称重算法研究  

About On-board Weighing Device System Optimization Design

作　　者：李惠敏[1] 李晓林[1] 

机构地区：[1]太原理工大学信息工程学院,山西太原030600

出　　处：《计算机仿真》2016年第11期140-143,共4页Computer Simulation

摘　　要：在车载动态称重优化的研究中,由于传统称重算法车与称重设备是分离的,造成车载动态称重数据精度较低。为此提出了利用小波变换和径向基神经网络(RBF网络)的车载称重算法。首先利用小波变换对称重信号滤波预处理,然后以称重信号、倾角信号、速度信号、加速度信号和车型组建RBF网络模型,并对其进行大量的实测数据训练,对训练之后的RBF网络模型进行车辆动态加载称重实验验证。实验结果表明,系统称重精度高、实用性强,能够满足当前的车载称重需求。Aiming at the problem of low accuracy of vehicle dynamic weighing data, a weighing data algorithm based on wavelet transform and RBF neural network （RBF） is proposed. Firstly, the weighing signal filter is preprocessed by using the wavelet transform. Secondly, a RBF network model is formed based on weighing signals, angle signal, speed and acceleration signals. Finally, a large number of experimental data are trained, and the RBF network model with vehicle dynamic load weighing is tested. The results show that the designed system has weighing accuracy and strong practicability, which can meet the demand of the current vehicle weighing.

关 键 词：车载称重 小波变换 径向基神经网络 倾角信号 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]