基于BP网络的整车式动态称重数据处理

Full-Vehicle Dynamic Weighing Data Processing Based on BP Algorithm

在整车称重准确性优化的研究中,整车式动态称重信号中的干扰信号,影响称重的准确性。由于在称重信号中存在大量的低频干扰和高频噪声,同时车辆通过称重区域时的行驶速度、加速度和轴型,均影响动态称重结果的正确性。针对以上问题,提出首先利用单子带重构改进算法对称重信号进行滤波与信号重构处理,然后利用BP神经网络算法在线自学习方法确立称重模型后,结合轴型、车速和加速度对称重信号进一步校正。利用上述算法进行仿真并应用于现场,使得称重数据的误差控制在0.5%以内,取得良好的称重效果。

The interference signals in vehicle dynamic weighing affect the weighing accuracy. There are a lot of low frequency noise and high frequency noise in the weighing signal. At the same time, the driving speed, acceleration and axial type of vehicle in weighing region also affect dynamic weighing results. This paper presents a full-vehicle dynamic weighing data processing algorithm based on BP Firstly, the weighing signal is filtered reconstructed by using the list with reconstruction improvement algorithm. Then the BP neural network is used to build the weighing model, and the weighing signal is further corrected combined with axial type, speed and acceleration. The results of simulation and the on-site application show that the proposed algorithm has good weighing performance, and the error of the weighing data is less than 0.5%.