Research On Single Gimbal Control Moment Gyroscope Servo Control Technology Of High Resolution Remote Sensing Satellite

High-resolution remote sensing satellites are widely used in the fields of environmental monitoring,agricultural production,geographic mapping,meteorological forecasting,resource exploration and modern military.By improving the agility of the remote sensing satellite attitude control system,the remote sensing satellite can complete multiple push-broom imaging tasks on the same target within the same orbital period and shorten the revisit period,thereby ensuring the stable and fast acquisition of high-quality images by the high-resolution remote sensing satellite,this point has always been one of the research focuses of high-resolution remote sensing satellites.The improvement of attitude control agility of remote sensing satellites relies on attitude control components that are capable of stable output of large torque.Compared with traditional satellite attitude control components such as jet components,flywheels and double gimbal control moment gyroscope,single gimbal control moment gyroscope have the advantages of high output moment,light weight,low power consumption,clean and non-polluting,no gimbal locking and high maneuverability.Since the output moment of single gimbal moment gyroscope is the product of gimbal angular velocity vector and flywheel angular momentum vector,and the flywheel output angular momentum is constant value.Therefore,the accuracy of the output torque of single gimbal control moment gyroscope depends entirely on the gimbal servo system control accuracy,and it is important to study the accuracy of single gimbal control moment gyroscope gimbal servo system to improve the output torque accuracy and thus the satellite attitude control accuracy.This paper is based on the “Control Moment Gyroscope” project of Chang Guang Satellite Technology Co.,LTD and“Youth Talent Support Project”,to design and study the single gimbal control moment gyroscope gimbal servo system used in high-resolution optical remote sensing satellites.In this paper,the research is carried out in several aspects such as motor control,sensor accuracy,system model and control algorithm.To improve the control accuracy of gimbal servo system,which is important to guide the design of control moment gyroscope system and ensure the system accuracy.The main research work is divided into four areas as follows.(1)The gimbal servo system motor model and control method are studied.First,the motor selection is performed according to the requirements of the single gimbal control moment gyroscope gimbal servo system.The motor structure and characteristics are introduced,and the mathematical model including voltage equation,torque equation,magnetic chain equation and motion equation is derived.Next,the coordinate transformation method for space vector control of the permanent magnet synchronous motor and the principle of pulse width modulation based on the vector control strategy are introduced.Finally,a simulation model of the gimbal servo system based on the PI algorithm is built in Matlab/Simulink to track the speed step signal and the sine signal,and according to the speed and current response signals,the effectiveness of the vector control strategy is proved.(2)The accuracy of the position sensor of the gimbal servo system is studied.Starting from improving the accuracy of the angular position sensor circle grating in the SGCMG servo system,a software compensation method using a single reading head plus a compensation algorithm is proposed to replace the use of a dual reading head hardware compensation method to eliminate eccentricity errors.First,analyze the output torque principle of the single gimbal control moment gyroscope system.The analysis results show that the measurement accuracy of the circular grating sensor will affect the output torque accuracy of the single gimbal control moment gyroscope.Secondly,analyze the source of the angle error of the circular grating,and derive the theoretical model of the tilt error and eccentric error in the installation error.The circular grating calibration experiment was carried out,and the compensation model was obtained according to the error data obtained from the experiment and the comprehensive analysis of the error geometric model.Finally,the software compensation algorithm is applied to compensate the angle data collected by the single reading head,which can increase the measurement error from 311.18 " to 6.23".The comparison shows that the acquisition accuracy of the circular grating can reach the compensation accuracy of the double reading head after using software compensation.(3)The gimbal servo system model is obtained by the method of system identification.Firstly,the single gimbal control moment gyroscope gimbal servo system is equated to the two-mass block model and the theoretical mathematical model of the system transfer function is derived.At the same time,a method to suppress the mechanical resonance frequency is proposed,and the mathematical model of the orthogonal correlation analysis method is introduced and analyzed.Secondly,the system model identification study is carried out,and the single gimbal control moment gyroscope is excited by the input sinusoidal sweeping current signal,and the output velocity signal is recorded synchronously to obtain the amplitude-frequency characteristics and phase-frequency characteristics of the system by orthogonal correlation analysis,and the transfer function of the system is obtained by fitting to the frequency characteristics.Finally,the frequency characteristic curve of the transfer function obtained by the fitting is compared with the frequency characteristic curve obtained by the experiment to verify the effectiveness and correctness of the identification algorithm.(4)The gimbal servo system control algorithm is proposed and simulations and experiments are conducted to verify its effectiveness.To address the problem of nonideal and non-linear disturbances in the servo system leading to the degradation of the system tracking performance,a fast terminal sliding mode control strategy with strong anti-interference capability and fast convergence is proposed,and a new convergence rate is designed to solve the jitter problem inherent in the sliding mode control.At the same time,a neural network adaptive PID control strategy is proposed.This control strategy can make the single gimbal control moment gyroscope have the function of onorbit parameter adjustment under the condition of occupying as little software and hardware resources as possible,and carry out software and hardware design to build an experimental platform.The algorithm proposed above is simulated and experimented to verify the control accuracy and dynamic accuracy of the algorithm.