针对BP基本算法应用于曲面重建的缺点,提出自调整S型函数方法,并利用共轭梯度法同时调整权值系数和S型函数的形状参数,使得收敛曲线能够迅速摆脱平缓的区域,加速了收敛进程。将这一改进的算法应用于BP网重建自由曲面中,仿真结果表明,同BP基本算法相比,在收敛速度和重建精度上均有较大程度的提高。 Aiming at the shortcomings of the BP algorithm applied to the surface reconstruction, a self-adjusting S-shaped function method is proposed and the shape parameters of the weighting coefficient and the S-shaped function are adjusted simultaneously by the conjugate gradient method, so that the convergence curve can quickly get out of the gentle area and accelerate Convergence process. The improved algorithm is applied to the BP network to reconstruct the freeform surface. The simulation results show that compared with the BP basic algorithm, both the convergence rate and the reconstruction accuracy are greatly improved.