Owing to the complicated variable rule of CMMs geometry error, it's difficult to convergence for using common BP neural network model arithmetic with a slow velocity.
由于坐标测量机几何误差变化规律复杂，采用一般的BP神经网络模型算法，速度慢且难以收敛。
Firstly, the error of fit must be defined for nonlinear least-square fitting of generalized geometry model. Then the nonlinear optimization algorithm can be used to obtain the optimum solution.
对于一般几何模型的非线性最小均方误差拟合，首先必须定义拟合误差，然后采用非线性最优化方法求解最小误差意义下的最优解。
应用推荐