论文部分内容阅读
针对车辆动态称重信号的特点,提出了一种车辆动态载重信号的处理方法。该方法采用基于神经网络的自适应滤波变步长LMS算法滤掉称重信号在各个频段内的噪声。在不同的环境下,对不同的车型,该技术适应性好,精度高,速度快。基于这种信号处理方法,开发了基于神经网络自适应滤波的车辆超载动态监测系统,选择高性能TMS32C2812芯片,设计了高效的软硬件系统,能够准确、及时测量高速公路上所经车辆的重量。
Aimed at the characteristics of vehicle dynamic weighing signal, a method of processing vehicle dynamic load signal is proposed. The method uses a neural network-based adaptive filtering variable step size LMS algorithm to filter out the noise of the weighing signal in each frequency band. In different environments, for different models, the technology adaptability, high precision, fast. Based on this signal processing method, a vehicle overloading dynamic monitoring system based on neural network adaptive filtering was developed. High performance TMS32C2812 chip was designed and efficient hardware and software system was designed to accurately and timely measure the weight of the vehicles on the highway.