Research On Adaptive Temperature Control Method For Inertial Platform System Based On Neural Network

The precision of gyro in inertial platform system is greatly influenced by temperature. It can be seen that the inertial platform temperature control system directly affects the precision of the inertial navigation system. Therefore, it is significant to improve the accuracy of the inertial navigation system to improve the precision of the inertial platform. In this paper, the heat transfer process of a certain type of high precision inertial platform is analyzed, in view of the complicated structure and multi coupling of the temperature control object, the mathematical simulation model of the system is established based on the correlation analysis method based on the correlation analysis method based on pseudo random number.Aiming at the existence of nonlinear link and disturbance in the temperature control loop, the prediction model of the system is established by using the method of genetic optimization neural network system identification. Combined with the single neuron adaptive PSD controller, the neural network adaptive intelligent controller is constructed, which can enhance the anti disturbance ability of the inertial platform temperature contents of this paper include the following aspects:First through the analysis of heat transfer process of a certain type of high precision platform inertial internal platform and inertial instrument, pushing structure of the relationship matrix between derived inertial platform and inertial instrument of heating power and the temperature. In order to simplify the calculation and simulation analysis, simplify the structure form of the relationship matrix. By studying the characteristics of the pseudo random number and the construction method, the input and output data are obtained. Based on the correlation analysis identification method based on pseudo random number, the parameters of the relationship matrix are identified. The mathematical model of the temperature control object of the inertial platform is established.Secondly, based on the solution of the nonlinear problem of temperature control loop and the complex angle of the model, the system prediction model with neural network structure is established by the method of neural network system identification. In order to improve the precision of neural network identification, the method of using genetic algorithm to optimize the initial parameters of neural network is adopted, and the prediction model of the inertial platform temperature control system is established.Finally, based on the traditional PID controller, a single neuron adaptive PSD controller is designed based on the two performance index in order to raise the anti disturbance ability of the temperature control system, Neural network adaptive intelligent controller based on neural network prediction model. The simulation and experimental results show that the control method can improve the anti disturbance ability of the inertial platform temperature control system, and improve the temperature control effect of the platform.