Abstract
Reservoir management is based on the prediction of reservoir performance by means of numerical simulation models. Reliable predictions require that the numerical model mimics the known production history of the field. Then, the numerical model is iteratively modified to match the production data with the simulation. This process is termed history matching (HM).
Mathematically, history matching can be seen as an optimisation problem where the target is to minimize an objective function quantifying the misfit between observed and simulated production data. One of the main problems in history matching is the choice of an effective parameterization: a set of reservoir properties that can be plausibly altered to get an history matched model. This issue is known as parameter identification problem and its solution usually represents a significant step toward the achievement of an history matched model
In this paper we propose a practical implementation of a multi-scale approach to identify effective parameterizations in reallife HM problems. The approach is based on the availability of gradient simulators, capable of providing the user with derivatives of the objective function with respect to the parameters at hand. Those derivatives can then be used in a multi-scale setting to define a sequence of richer and richer parameterisations. At each step of the sequence the matching of the production data is improved. The methodology validated on synthetic case and has been applied to history match the simulation model of a North-Sea oil reservoir.
The proposed methodology can be considered a practical solution for parameter identification problems in many real cases. This until sound methodologies, primarily adaptive multi scale estimation of parameters, will become available in commercial software programs.
