基于径向基神经网络的速度滤波器  

Velocity filter based on RBF neural network

作　　者：孙丽[1] 孙铁利[1] 张建国[2] 

机构地区：[1]东北师范大学计算机学院,吉林长春130117 [2]中国人民解放军装甲兵技术学院,吉林长春130117

出　　处：《计算机工程与设计》2008年第4期904-905,908,共3页Computer Engineering and Design

摘　　要：在实车采集的试验数据当中,速度信号对于分析车辆的整车性能、换档过程以及通过数据处理进一步求取加速度、冲击度等指标有着非常重要的作用。但是由于试验当中各种随机噪声的存在,因此必须在研究分析之前先对速度信号进行滤波。径向基神经网络能够任意逼近非线性函数的特性,给速度滤波提供了崭新的方法。利用径向基神经网络可实现自适应滤波,并应用于速度信号的去噪。仿真结果表明,该方法具有良好的非线性噪声抑制能力。Velocity signal as one of the important parts in the synchronized data acquisition is help to analyze the whole vehicle performance and shift process. The acceleration and jerk signal can be easily attained through the processed velocity data. However there is a lot of noise in this signal. So the velocity signal must be filtrated before analysis. Radial basis function neural network can approximate every nonlinear function, so it can be used to filtrate the velocity signal. A new filter based on RBF neural network is introduced. The simulation results indicate that this filter has good performance in controlling signal. And the noise can be removed efficiently.

关 键 词：神经网络 车辆工程 自适应滤波 非线性函数 速度滤波器 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]