基于遗传算法的低轨卫星轨道预报伪随机脉冲参数优化OA

To optimize the pseudo-random pulse parameters in reduced-dynamic orbit determination,a genetic algorithm-based optimization strategy for pseudo-random pulse parameters is proposed.The applicability of this strategy is analyzed using GRACE-FO satellites,and its effectiveness is validated by comparing it with the traditional empirical pulse parameter determination method as a benchmark.The results show that the proposed strategy improves the orbit determination accuracy for the GRACE-C and GRACE-D satellites by up to 38.5%compared to traditional methods,achieving a peak single-day accuracy of 11.5 mm.In terms of orbit prediction,the optimal accuracies for 1 h,2 h,and 3 h predictions are 27.9 mm,52.1 mm,and 72.7 mm,respectively.The 3 h prediction accuracy for GRACE-C and GRACE-D improves by an average of 46%and 23%,respectively.The strategy yields the most significant error improvement in the A-direction,while its impact on the C-direction and R-direction is relatively smaller.These findings provide a reference for improving both the accuracy of orbit determination and prediction.