Research On Bias Temperature Characteristics And Compensation Technology Of Micro-hemispherical Gyroscope

Micro-hemispherical gyroscope is a research hotspot of MEMS gyroscope research in-stitutions at home and abroad in recent years.Compared with traditional MEMS gyroscope,micro-hemispherical gyroscope is more suitable for high precision occasions.However,the mechanical structure error of the micro-hemispherical gyroscope is more sensitive to the envi-ronment.When the ambient temperature changes,it is easy to cause bias drift.This article takes micro-hemispherical gyroscope as the research object,and studies the temperature characteristics of its bias error.The main research contents include:(1)Analysis of working principle of micro-hemispherical gyroscope.Mainly discussed the structural characteristics,working mode,excitation and detection method of the micro-hemispherical gyroscope.Analyzed the dynamic equations of the micro-hemispherical gyro-scope’s force balance working mode,and introduced the control system in the micro-hemispherical gyroscope in force feedback mode.(2)Research on bias drift mechanism of micro-hemispherical gyroscope.The causes of the bias error of the micro-hemispheric resonator are analyzed from two aspects:the structure error of the resonator and the control system error.The structure error of the resonator mainly includes the damping coupling error and the stiffness coupling error.The control system error mainly includes the circuit feedthrough error and the phase modulation angle error.Then the influence of temperature on the resonant frequency,Q value of the micro-hemispherical resonator and the control circuit are studied.Among them,the temperature and the resonant frequency of the micro-hemispherical resonator are basically linearly positively correlated.Reflect changes in ambient temperature.Finally,the influence of temperature on the bias error of the resonator is analyzed by integrating various factors:the complex nonlinear relationship between the bias error of the resonator and the temperature.(3)Research on the compensation method of gyro bias drift based on bias-temperature drift model.It mainly introduces three model fitting methods:gyro bias drift compensation method based on piecewise polynomial fitting,gyro bias drift compensation method based on BP neural network,and bias drift compensation method based on particle swarm optimization to optimize BP neural network.The bias drift of the micro-hemispherical gyroscope is compensated by fitting the bias-temperature drift curve.The bias drift compensation method of gyroscope based on piecewise polynomial fitting is mainly to manually segment the bias-temperature drift curve,and to fit each piece separately with different low-order polynomials.The gyro bias drift compensation method based on BP neural network fits the gyro bias drift characteristic curve through the BP neural network model.The bias drift compensation method based on particle swarm optimization to optimize the BP neural network is an optimized version of the BP neural network model proposed to solve the problem that the ordinary BP neural network model is easy to fall into the local minimum during the training process.It is easier to find the overall optimal solution.(4)Research on bias-drift compensation system of micro-hemispherical gyroscope based on dual-mode actuation.The working principle of the dual-mode actuation gyroscope is analyzed by formula derivation:the two axial directions of the micro-hemisphere are simultaneously excited.The displacement sum of the vibrations of the two axial directions is the driving mode of the gyroscope.The displacement difference of the two axial vibrations is the gyroscope The detection mode of the instrument.Based on the principle simulation,the specific control mechanism of the dual-mode excitation gyro drive detection is studied:the harmonic drive is controlled at the center frequency of the two axial resonance frequencies by the in-phase demodulation signals of the sum of the two axial vibration displacements.The driving amplitude is controlled by the quadrature demodulation signal.The in-phase demodulation signal of two axial vibration displacement differences is the output signal of the resonance frequency and the quadrature demodulation signal is the Coriolis output signal.Then the mechanism of dual-mode actuation gyroscope tuning control and the mechanism of Q value mismatch compensation are studied.Finally,the measurement and control block diagram of the bias-drift compensation system of the micro hemispherical resonant gyroscope based on dual-mode actuation is given.(5)Compensation effect test of micro-hemispherical gyroscope.Refer to the test spec-ifications of the domestic micro-machined gyroscope to test the performance of the micro-hemispherical gyroscope before and after compensation.The experimental results show that after compensation,the bias-drift compensation method of the micro-hemispherical gyroscope based on dual-mode actuation reduces the value of the bias-temperature stability of the micro-hemispherical gyroscope at room temperature by 4 times,from single mode actuation of 227~°/h is reduced to about 53°/h,the angle random walk is reduced by about 2 times,from single mode actuation of 34°/h/~√Hz reduced to 15°/h/~√Hz.However,the compensation effect of the gyro bias drift compensation method based on the bias-temperature drift model is relatively poor.Among them,the bias drift compensation method based on piecewise polynomial fitting reduces the room temperature bias stability value by 1.5 times.And the bias drift compensation method based on BP neural network reduces the room temperature bias value by 2 times.Based on PSO-BP neural The bias-drift compensation method of the network reduces the bias stability value by 2.5 times,and has no compensation effect for the random walk of the angle.From the analysis of the repeatability of the compensation effect,the bias-drift compensation method based on dual-mode actuation has little overall fluctuation in the compensation effect,and the repeatability is better.