Problems of Identification of Hydrodynamic Models in Reservoir Engineering

The need for solving the problem of identification of basic parameters (permeability coefficients, initial and boundary conditions) for hydrodynamic models of reservoirs on retrospective data arises from incompleteness and observational errors in source and current information about controlled object at all stages of reservoir engineering. Iterative methods for simultaneous solving of direct and inverse problems for original model equations on retrospective data are proposed to solve the problems of identification and adaptation of initial and boundary conditions and filtration parameters. Parallel computing technologies with optimal hierarchical (multilevel) embedding of algorithms into the architecture of supercomputers are offered for creation an integrated model of complex technological system.