Research On A Precise Compensation Method For Temperature Drift Error Of IFOG

As the core component of a new strap-down INS,Fiber Optic Gyroscope(FOG),its optical components and electrical components have an outstanding temperature dependency,which causes FOG very sensitive to the ambient temperature.Meanwhile,it also remarkably reduces the accuracy of strap-down INS.Hence,researching a precise compensation method for temperature drift error of FOG is very important to improve the temperature dependency for the external and the internal environment and guarantee the accuracy of strap-down INS.This thesis takes marine inertial navigation system based on FOG as the research background and the laboratory investigation FOG as the study object,and researches a precise compensation method for temperature drift error of FOG.The research purpose is to improve the environmental adaptability of FOG,and guarantees FOG to output the angular velocity of the carriers accurately,stably and in real-time.The work is mainly shown as followed:Firstly,the research background and significance are demonstrated,and the development status of FOG,temperature control for FOG and temperature compensation for temperature drift error of FOG are investigated and analyzed.General requirements and key technologies for a precise compensation method for temperature drift error of FOG are shaped.Then,it proposes the overall scheme design of a precise compensation method for temperature drift error of FOG based on temperature control and temperature compensation.Secondly,in order to increase the accuracy of temperature measurement,a high-accuracy temperature measuring method based on sequential driving voltage control is researched and proposed.With the assistance of the temperature sensor Pt1000,resistance comparison is researched and introduced to eliminate the nonlinear errors in temperature measurement,and sequential driving control is researched and introduced to remove thermal electromotive force and to restrain the self-heating effect of Pt1000.Then,AD synchronous sampling method is researched and proposed to eliminate the AD sampling error due to the instability of the driving voltage and the reference voltage.The parameters of the method based on sequential driving control is designed and optimized.Then,a correction method is designed with piecewise linear fitting to reduce the fitting error and enhance the calculating real-time.According to the sampling characteristics,a low-pass butterworth filter is designed to eliminate the random error.Finally,the high-accuracy temperature measuring method is tested for a long time with a thermostat.Thirdly,aiming at solving the problem of the poor environmental adaptability of FOG,on the basis of illustrating how smith predictor improves the control performance of PID temperature control system,it researches and proposes small temperature variation control method based on Smith predictor.Aiming at dealing with the problem that the performance of Smith predictor reduces in some situations,by simulating and analyzing the influence of the parameters mismatch between smith predictor and the control object to the system performance,a discrete approximation estimation model(DA model)for the thermal chamber is designed and established according to the point of signal sampling characteristics of digital control system.The parameters of the thermal chamber is identified with the temperature samples from PID temperature control system precisely,and DA model is built to estimate the parameters of the thermal chamber of PID temperature control system accurately and in real-time to update the parameters of Smith predictor.Meanwhile,the output of PID controller is adjusted to keep temperature inside the thermal chamber varying in a small temperature variation which stabilizes at the target accurately and smoothly.Then,a temperature experiment is designed.According to the test results,the parameter estimation performance of DA model and Fuzzy model are compared and analyzed from the dynamic characteristic and the steady characteristic.Additionally,the performance of small temperature variation control is analyzed.Fourthly,to solve the problem that optical fiber ring have the remarkable temperature dependency,the causes for temperature drift error of FOG are analyzed,and an important factor to temperature drift error of FOG is deduced,temperature combined term.The traditional estimation model for temperature drift error of FOG is reshaped with temperature combined term,and a modified estimation model for temperature drift error of FOG is established with temperature of fiber loops,temperature variation of fiber loops and temperature combined term of fiber loops.Then,the laboratory investigation FOG is tested in the temperature experiments.According to the test results,a modified estimation model for temperature drift error of FOG is implemented based on Radial Basis Function Artificial Neural Network(RBF-ANN)with temperature related term of fiber loops and temperature drift error of FOG.Then,a temperature experiment is designed to test the traditional estimation model for temperature drift error of FOG and the modified estimation model for temperature drift error of FOG,and their performance are compared and analyzed.At last,in order to test the performance of precise compensation methods for temperature drift error of IFOG,considering the accuracy and stability of angular velocity from FOG as an examination method,compensation methods for temperature drift error of FOG respectively based on temperature control,temperature compensation,temperature control and temperaturecompensation are tested in terms of performance,and their performance are compared and analyzed.The results show that after being compensated with the compensation methods for temperature drift error of FOG based on small temperature variation control method and the modified estimation model,the variation of temperature inside the thermal chamber decreases obviously to about 60% of the primary value,and the overshoot of PID temperature control system is eliminated.Meanwhile,the accuracy of FOG is enhanced to ±0.05°/h,and the mean square deviation of FOG after compensating is about two orders of magnitude higher than that before compensating,improving to 1.932% of its primary value.Therefore,the precise compensation method for temperature drift error of FOG based on small temperature variation control method and the modified estimation model can reduce the variation of temperature inside the thermal chamber effectively to enhance the environmental adaptability of FOG,and maintains the temperature inside the thermal chamber stable to improve the accuracy and real-time of FOG.More importantly,the method can estimate and compensate temperature drift error of FOG accurately to decouple the temperature dependence of the fiber loops of FOG effectively and guarantee that FOG can output stably and precisely.To ensure INS run accurately,stably and reliably in different working conditions,a precise compensation method for temperature drift error of FOG based on small temperature variation control method and the modified estimation model is of great significance.