基于最优加权平均模型的高程异常拟合分析

Elevation Anomaly Fitting Analysis Based on the Optimal Weighted Average Model

单一数学模型在似大地水准面精化中应用广泛,适用于特定的地势地区,但精度难以符合要求,且对局部地势复杂地区无法准确拟合。针对这一情况提出了对某地区GPS/水准数据分别利用多种单一数学模型进行拟合,并对结果通过一种最优加权平均的方法进行组合。通过计算内外符合精度、检核点最大最小误差、检核点误差小于1 cm、2 cm、3 cm、4 cm的个数,结果显示,在地势较为平缓的地区,最优加权平均模型可以整体提高精度6%~28%,在实践中具有很好的拟合效果和实际意义。

The single mathematical model is widely used in geoid like surface refinement and is suitable for specific terrain areas,but its accuracy is difficult to meet the requirements,and it cannot accurately fit the local terrain complex areas.In view of this situation,several single mathematical models were used to fit the GPS/leveling data in a certain area,and the results were combined by an optimal weighted average method.By calculating the number of internal and external coincidence accuracy,maximum and minimum error of inspection point,and error of inspection point less than 1 cm,2 cm,3 cm and 4 cm,the results show that in the area with relatively gentle terrain,the optimal weighted average model can improve the overall accuracy by 6%~28%,which has a good fitting effect and practical significance in practice.