论文部分内容阅读
采用数据融合方法研究了基于多地震波传感器的管道安全监测预警系统中目标特征提取及识别问题。用多个传感器及处理模块采集管道周围目标产生的震动信号,采用经验模态分解方法对地面震动信号进行处理,提取分解结果的归一化峭度,将其作为特征向量;利用主要频率区间的特征向量进行单一传感器的目标识别。由于监测系统由多个单独传感器采集模块组成,为了提高目标的识别率,采用D-S证据理论对识别结果进行了数据融合,得到最终识别结果。利用该方法对实验采集数据进行处理,验证了文中提出的方法。
The data fusion method is used to study the target feature extraction and identification in pipeline safety monitoring and early warning system based on multi-seismic sensors. A plurality of sensors and processing modules are used to collect the vibration signals generated by the target around the pipeline. The ground-based vibration signals are processed by the empirical mode decomposition method, the normalized kurtosis of the decomposition results is extracted and used as the eigenvector. By using the main frequency range Eigenvector for single sensor target recognition. Because the monitoring system is composed of a number of separate sensor acquisition modules, in order to improve the recognition rate of the target, the recognition results are fused using D-S evidence theory to obtain the final recognition result. Using this method to deal with the experiment data, we verify the proposed method.