Detection method of laser gyroscope cavity adjustment based on momentum BP neural network

In order to solve the manual detection drawbacks of laser gyroscope cavity adjustment, such as low quality and low efficiency, a multi-sensor information fusion detection method is proposed using a CCD camera and a photomultiplier. The center of the facula and the diaphragm and the loss of laser gyroscope are obtained and then transmitted to the fusion center. After fusion calculation, the integrated judgment is produced. The fusion system utilizes the momentum back-propagation neural network (BPNN) to fuse the multi-source information and output the final decision. And according to the modes of the detected signals and output decision, a three layers topology structure including an input layer, a hidden layer and an output layer is designed. The experimental results indicate that the accuracy of the proposed cavity adjustment detection method is 93.81%, which is higher than the manual step detection method using a single sensor about 6%.