作者: Clayton V. Deutsch , Perry W. Cockerham
DOI: 10.1007/BF02065876
关键词: Numerical models 、 Perturbation (astronomy) 、 Stochastic modelling 、 Simulated annealing 、 Algorithm 、 Computer science 、 Conditional simulation 、 Stochastic simulation 、 Optimization methods 、 Optimization problem 、 Mathematical optimization
摘要: Realizations generated by conditional simulation techniques must honor as much data possible to be reliable numerical models of the attribute under study. The application optimization methods such simulated annealing stochastic has potential more than conventional geostatistical techniques. essential feature this approach is formulation imaging an problem with some specified objective function. honored images are coded components in a global This paper describes basic algorithm and then addresses number practical questions: (1) what criteria for adding component function? (2) perturbation mechanism should employed simulation? (3) when temperature lowered procedure? (4) how edge/border nodes handled? (5) local conditioning (6) multiple weighted