论文部分内容阅读
为解决激光陀螺人工调腔质量低、效率不高等缺点,提出一种由CCD相机和光电倍增管构成的多传感器信息融合调腔检测方法,该方法同时检测光斑、光阑中心点及陀螺损耗值,并将这些信息送入融合中心,经过融合计算得到陀螺调腔是否合格的综合判断。融合系统采用动量BP神经网络对多源信息进行融合,根据调腔过程中的输入、输出情况,设计包含输入层、隐含层及输出层的3层网络拓扑结构。实验结果表明,此种方法对激光陀螺调腔质量是否合格判断准确率为93.81%,比人工采用单一传感器分步检测准确率提高了6%。
In order to solve the problem of low quality and low efficiency of artificial cavity in laser gyro, this paper proposes a multi-sensor information fusion cavity detection method which is composed of CCD camera and photomultiplier tube. The method detects flare, aperture center and gyro loss simultaneously , And these information into the fusion center, after fusion calculation gyroscopic cavity is qualified to judge. The fusion system uses the momentum BP neural network to fuse multi-source information. According to the input and output conditions in the process of cavity tuning, a 3-layer network topology including input layer, hidden layer and output layer is designed. Experimental results show that the accuracy of this method for determining the quality of laser gyroscope cavity is 93.81%, which is 6% higher than that of manual detection by single sensor.