| 主权项 |
1. A method for automated calibration of a plurality of stratigraphic models to available geological data, such as geological prior information, well logs and/or seismic data, and any uncertainties relating to the available geological data, the method comprising the steps of:
a) providing a process-based Stratigraphic Forward Modeling (SFM) tool that generates a stratigraphic model of the geological evolution and fill of a sedimentary basin on the basis of a set of input tectonic and sedimentary process parameters; b) providing a conditional sampling algorithm that tests different sets of input process parameters to the SFM tool, by sampling neighborhoods provided by generalized hypervolumes of those input sedimentary process parameter sets that yield stratigraphic models with best agreement to the available geological data; c) providing a local minima explicit escape clause of the iterative conditional sampling algorithm, which escape clause identifies local minima that are defined as models in a densely sampled region of the parameter space with an agreement to calibration data that does not meet the error threshold criteria to be admissible solutions, yet exhibit a lower error than neighbouring models, and forbids the oversampling of said minima, thereby allowing the sampling algorithm to contine the search in other regions of probable solutions; d) providing an admissible solution explicit escape clause of the iterative conditional sampling algorithm, which identifies admissible solutions that are defined as models with an agreement to calibration data that meet a predefined error threshold, and which then prevents the algorithm from oversampling said admissible solutions, thereby allowing the SFM tool to continue a search in other regions of probable solutions; and e) iterating steps b, c and d a number of times in order to identify a plurality of admissible stratigraphic models that are calibrated to the available geological data. |
