提出了一种求解portial errors-in-variables(partial EIV)模型的思路。通过对partial EIV模型的部分元素进行移项,重组成新形式下的平差函数模型,两次运用间接平差原理分别求解平差参数与系数矩阵中的随机元素,把总体最小二乘平差问题转化为最小二乘平差问题,并通过适当变换提高了新解法的收敛速度。最后分别采用实测数据和模拟数据进行验证,求解了本文算法与已有算法的估值结果。算例结果表明,本文算法能取得与已有算法相同的结果,是切实可行的。
A new thinking for solving partial errors-in-variables (partial EIV) model was proposed. Through the transposition processing in partial EIV model, a new functional model was reconstructed. Adjustment of indirect observations has been used two times to calculate the model parameters and the stochastic elements in coefficient matrix, translating total least squares problem to least squares problem. It also achieves high convergence rate through some simple variables transformation. Finally, real and simulation data were implemented to compare with the existing algorithms and to analysis the applicability of the proposed algorithms. The results show that the new algorithms are feasible and it can achieve the same values with the existing algorithms.