根据国内大量的地表移动观测资料,建立了地表移动盆边界角计算的神经网络模型.该模型与遗传算法相结合,克服了单一神经网络易陷入局部最优和收敛速度慢等缺点。研究表明,用基于遗传算法的神经网络来计算地表移动盆地的边界角,其结果更加真实、可靠。 According to a large number of observation data of surface movement in China, a neural network model for calculating the boundary angle of the mobile basin is established, which combines with the genetic algorithm to overcome the shortcoming of single neural network being trapped in local optimum and slow convergence. The research shows that the neural network based on genetic algorithm is used to calculate the boundary angle of the surface moving basin. The result is more realistic and reliable.