HGA: A genetic algorithm method for direct estimation of paleostress states from heterogeneous fault-slip data

摘要: Abstract Field data on fault-slip observations is commonly heterogeneous. Paleostress estimation from such sets is, in general, carried out two steps: (i) the classification of heterogeneous set into homogeneous subsets and (ii) an inversion each subset. This study gives a new approach, HGA, that combines issues single step process stress tensors directly. The given are directly operated upon by genetic algorithm operators, initialization, elitism, selection, encoding, crossover mutation. These operations simulate guided search finds successively fitter solutions, tensors, until globally fittest solution obtained. We first explain basic steps working example then demonstrate its veracity using several synthetic natural examples. proposed method obviates necessity having to classify sets. It estimates different states data. In contrast existing linear methods, not vulnerable entrapment local optimum. Although requires priori estimate maximum number expected population, this does control final results. Like any other method, too has merits limitations these discussed.