A multi-well data inversion purpose-built for calibration of an effective basin model at overpressure prediction

The calibration of a basin scale hydro-dynamic model based on a multi-well data set has been introduced as an ill-posed inversion problem. Global and local optimisation strategies have been reviewed in this connection. In terms of the conditionally unique definition according to Tikhonov this problem is getting a range of locally unique ε-equivalent solutions. All of them are a potentially resolve calibration problem. It is shown that for the multi-well case the relevant inverse problem solution can be arranged with additional criteria, which provide globalisation and lead to the area unique solution of the calibration problem. The approach, strategy and computation algorithm have been developed. The approach implies inverting of porosity - pore pressure data sets attached to the calibration well positions within the range of established and upscaled in the advance effective basin model. The numerical implementation has been checked on theoretical and real data examples. It reveals acceptable converging speed at practically justified data quality stop criterion. The application of the developed algorithm and numerical scheme for the 12-wells real data example (the North Sea) has been discussed in details.