A novel method for matching theoretical variogram based on particle swarm optimization
基于粒子群优化的理论变异函数拟合方法研究

Variogram is the main analysis tool of regionalized variable spatial structure and spatial local interpolation in geostatistics. In particular, theoretical variogram model is essential in geostatistics, which is the key issue in investigating the variability of regionalized variable and further geostatistics computing. Concerning the disadvantages of conventional methods that match theoretical variogram such as man-made matching method, linear programming method, weighted polynomial matching and goal programming method, a novel method for matching theoretical variogram based on particle swarm optimization was presented allowing for the characteristic of global optimization of particle swarm optimization algorithm. The experimental results show that the proposed method has higher prediction precision and stronger robustness than weighted polynomial matching based approach.