基于PSO算法的变差函数球状模型参数拟合

对一阶变差函数球状模型及其二阶套合结构的参数拟合进行研究，利用粒子群优化(PSO)算法在求解非线性优化问题时收敛的快速性以及全局寻优的有效性等优势，将待拟合球状模型的参数组合为一个粒子向量，在PSO算法迭代过程中对部分粒子进行混合柯西-高斯变异，实现变差函数球状模型最优参数的自动拟合。仿真实验结果表明，该方法操作简单、可靠性高。

Parameter Fitting of Variogram Spherical Model Based on Particle Swarm Optimization Algorithm

This paper researches the regression of the first spherical model and its mugget structure of theoretic variogram, and takes advantages of the Particle Swarm Optimization（PSO） algorithm which has outstanding advantages in solving nonlinear optimization problems for the rapid convergence and the effectiveness of global optimization. Its combination of the parameters in the spherical model are considered as a vector, during the interation of PSO algorithm, and mixed Cauchy-Gaussian mutation is carried out on part of the particles. The automtic regression of the spherical model is achieved. Simulation results show that this method is simple and has high reliability.