Fiber Optic Gyroscope Based On RBF Neural Network-Research And Application Of Temperature Compensation System

Fiber optic gyroscope is an inertial component based on the Sagnac effect,which detects the phase difference of interference light to achieve angular velocity measurement.Fiber optic gyroscopes can be applied in many situations.By measuring the movement posture of armored vehicles using fiber optic gyroscopes,it can ensure that they can still accurately aim and track the attacking target during bumpy travel;The track detection system based on fiber optic gyroscope can depict the three-dimensional trajectory of the track,which can detect key parameters such as track gauge,direction,and slope,ensuring the safe operation of high-speed trains and subways.The output of fiber optic gyroscopes is very sensitive to temperature changes.In engineering practice where temperature is constantly changing,mathematical modeling is usually used to identify the relationship between environmental temperature and the zero position output of fiber optic gyroscopes,establish a temperature model,compensate for the output,and reduce temperature dependence.Artificial neural network is a parallel computing system composed of a large number of interconnected neurons,which has been widely applied in various fields such as signal processing and medical treatment.The working environment temperature and angular velocity output of fiber optic gyroscopes are a nonlinear relationship.A neural network temperature compensation system based on FPGA(Field Programmable Gate Array)can not only improve processing speed,but also save hardware resources and reduce volume.The research content and innovative points of this article are as follows:1.According to the working principle of fiber optic gyroscope,design the circuit and parameters to meet the closed-loop requirements of fiber optic gyroscope.To obtain a stable output fiber optic gyroscope,a set of fiber optic gyroscope temperature compensation system is designed,and corresponding modules are designed according to the system requirements.2.According to the principle of RBF(Radial basis function)network,the RBF temperature compensation model is established by collecting the ambient temperature and the output angular velocity,and the algorithm is used to optimize the parameters of RBF network.At the same time,the particle swarm optimization algorithm is improved,and the Blending inheritance algorithm is used to optimize the particle swarm optimization algorithm in the process of population crossover and mutation as well as the Simulated annealing process,so that the RBF neural network temperature compensation model has higher accuracy.3.Provide an implementation scheme for RBF network on FPGA.According to the requirements of the designed temperature compensation system,design and write the modules required in FPGA,including temperature measurement,data conversion,and communication modules.4.The assembly and debugging,temperature compensation experiments,and accuracy analysis of the designed fiber optic gyroscope system have proven that the fiber optic gyroscope system designed in this article can work normally,and the established temperature compensation model can indeed reduce the temperature dependence of the fiber optic gyroscope.And compared with the temperature model established by traditional least squares fitting,the results show that the RBF temperature compensation model has better accuracy improvement and meets engineering needs.