摘要
波形数字声发射技术的发展给声发射源的特性识别带来了可能性。由于各种噪声的影响，以及声发射信号传播过程的复杂性，又给声源的识别带来一定的困难。为了解决干扰情况下声发射源的定性问题，文章提出了在决策层上的多传感器数据融合的识别方法。利用定位传感器组中各个传感器得到的数据，同时考虑在同一个定位组中各个传感器所得数据的置信度不同，对声发射源的性质进行识别。实验结果证明了数据融合后，声发射源特性识别的可靠性明显大于单个传感器的识别效果，这也表明了多传感器融合识别的可能性和有效性。
With the development of digital and wave-form AE (acoustic emission) technology, it makes the recognition ofAE source possible. Because of the noise and complex of AE signal transmission characteristic, the recognition of AE source isvery difficult. In order to resove this problem, the method of multi-sensor data fusion on decision level is presented. Using thedata from each sensor in a location set and considering the confidence value for each sensor, the AE source characteristics can beacquired. Experiment results show that the data fusion result is better than that of single sensor. The multi-sensor data fusionmethod is effective.
出处
《计算机测量与控制》
CSCD
2002年第5期345-348,共4页
Computer Measurement &Control
关键词
声发射源
多传感器
数据融合
识别
acoustic emission
data fusion
object recognition
