论文部分内容阅读
通过外场实验获得关于轮式车、履带式车的大量地震动信号,在时-频域应用多种方法对信号进行处理,得到相应的特征矢量。利用改进的BP网络对远距离的地震动信号进行目标识别,基于小波及小波包分解能量分布特征的识别率可达85%以上,这种特征矢量具有较好的可分性。
A large number of ground motions signals about wheeled vehicles and tracked vehicles are obtained through field experiments. Various signals are processed in the time-frequency domain to obtain the corresponding eigenvectors. The improved BP neural network is used to identify long-range ground motion signals. The recognition rate of energy distribution based on wavelet and wavelet packet decomposition is more than 85%. This feature vector has better separability.