Capturing Geological Realism in Stochastic Simulations of Rock Systems with Markov Statistics and Simulated Annealing

摘要: ABSTRACT Simulated annealing is a numerical algorithm that can be used to impose statistical structures on grids representing heterogeneous rock or sediment. In this paper, we use the flexibility of simulated generate with Markov structures. Our purpose transmit rich geological information captured in statistics into stochastic while maintaining honor field data. Performance issues compromise imbued Markovian properties include scales bedding bodies relative grid size, and amount complexity embedded The remedies these proper selection careful choice type, consideration an alternative stopping rule based chi-squared test statistic. If performance are overcome, complex stratal patterns such as higher-order dependency, cyclicity, directionality replicated by method. addition, accounting for variations depositional rate allows transference obtained from vertical boreholes horizontal dimension when other lacking. A example using borehole data collected at Gloucester special waste site near Ottawa, Canada, well synthetic examples, demonstrate technique issues.