4D Assisted Seismic History Matching Using a Differential Evolution Algorithm at the Harding South Field

Abstract

Summary 4D Assisted Seismic History Matching (4D ASHM) has been implemented and successfully applied to the Harding South field in the North Sea. A Success-History Based Parameter Adaption Differential Evolutionary (SHADE) algorithm was used to minimise an objective function derived from the observed and simulated 4D seismic data. Quantitative misfit values for the objective function were computed by binarisation of 4D attribute maps extracted from the observed and simulated 4D difference volumes. Multipliers of several reservoir model parameters including net-to-gross ratio, porosity, permeability and fault transmissibility were automatically updated through fifteen genetic evolutions with ten individuals in each generation. Reservoir simulations were run for each individual's model parameters and the property grids used to compute saturations, impedances and synthetic 4D seismic volumes through production time. The 4D ASHM process converged to stable solutions after 15 genetic evolutions. The objective function reached a minimum value with low variance and the the seven reservoir parameters reached stable values. The net-to-gross ratio and porosity were increased to provide a larger oil volume. The match between the observed and modelled 4D seismic data improved and the history-match to the producing wells was significantly better.