Research On Micro-and Nano-Satellite Magnetic Field Modeling And On-orbit Calibration Technology

Today,with the maturity and development of satellite technology,micro-and nanosatellites have attracted the attention of many research institutions and universities because of their low cost and short period.However,due to the limitations of physical size,weight,power consumption,and development cost,the micro-and nano-satellites cannot use the traditional satellite demagnetization means.The remanence becomes the maximum interference of the micro-and nano-satellite attitude determination and control accuracy.In order to better meet the development needs of micro-and nano-satellites and improve the overall performance of micro-and nano-satellites,this paper studies the magnetic field modeling and on-orbit calibration technology of micro-and nano-satellites.This has important engineering reference value.In the pre-design of micro-and nano-satellites,this paper establishes the model of residual magnetic field and residual magnetic moment of the whole star through the assumption of magnetic dipoles of micro-and nano-satellites and micro-nano satellite components,and adds the remanence as the optimization target to the satellite layout optimization design(SLOD)for the first time.The accelerated particle swarm optimization(APSO)algorithm is improved and used to solve the SLOD problem with the residual magnetic target.The simulation results show that adding residual magnetism to the SLOD problem can significantly reduce the residual magnetic field and residual magnetic moment in the pre-design of the micro-and nano-satellite.The optimized layout scheme obtained has guiding significance for the engineer layout.For the remanence calibration,firstly,before the micro-and nano-satellite launch,the residual magnetic field calibration technique based on ellipsoid fitting has the problem that the fixed magnetic field and the induced magnetic field coefficient cannot be separated.This paper proposes its own improvement measures and adopts the improved method.The method can successfully calibrate the fixed magnetic field and the induced magnetic field coefficient,and the error after compensation is the same magnitude as the random error of the magnetometer measurement.Then,the traditional method can not cope with the problem that the change of the condition of the micro-and nano-satellite in orbit operation has a great influence on the magnetic field environment of the micro-and nano-satellite.It is proposed to use the back propagation(BP)neural network algorithm to calibrate the residual magnetic field of the micro-and nano-satellite.The simulation results show that the calibration compensation accuracy is high,and it is suitable for the residual magnetic field of the micro-and nano-satellite to be calibrated in the orbit.Finally,for the problem of slow convergence of the remanent magnetic moment of microand nano-satellites by unscented kalman filter(UKF),two improvement strategies are proposed: one is to add solar sensitivity observations to the UKF observation equation;the other is proposed to apply the residual magnetic field calibration method based on ellipsoid fitting to calibrate the fixed magnetic field before the three-axis stability of micro-and nano-satellites,and verify its feasibility by simulation.Under the assumption of magnetic dipole,the residual magnetic moment of the micro-and nano-satellite is estimated by the calibrated fixed magnetic field strength.This is used as the initial value of the residual magnetic moment UKF to improve the filtering speed.Finally,this paper designs a ground magnetic measurement system and initially completes the physical production.This magnetic measurement system can measure the residual magnetic moment and the induced magnetic moment of the satellite component or the whole star.By measuring the magnetic torque component that has been calibrated,The correctness of the magnetic dipole hypothesis and the engineering practicality of the magnetic measurement system are verified.