Treatment of faults in production simulation models

Abstract

Abstract This paper describes basic ‘rules-of-thumb’ that offer an indication of common uncertainties and pitfalls, as well as the analytical methods, data requirements and work elements required to replicate the impact of faults on fluid flow in production simulation models successfully. The first, and most important, stage in this modelling process is to ensure that an accurate structural interpretation is incorporated into the simulation model. In particular, that all fault linkages and cross-fault juxtapositions are taken from the seismic interpretation into the simulation grid. Fault rocks sometimes reduce the rate of cross-fault flow in which case it is important to account for this reduction in flow within simulation models. This is best achieved if databases of fault rock properties, measured from the field of interest or nearby similar reservoirs, are up-scaled to calculate fault transmissibility multipliers. It is sometimes necessary to consider not just the single-phase permeability but also the capillary pressure and relative permeability characteristics of the fault rocks present. Finally, all the relevant static and dynamic data must be appraised critically. However, the interpretation of such data is usually non-unique and misinterpretations can create errors in the production-related fault seal analysis. Where these basic guidelines are followed, it has been our experience that the project time required to achieve a history match of production data is dramatically reduced. In addition, as the history match is more geologically reasonable, the model is often more reliable for predicting the long-term behaviour of the reservoir. This gives confidence in the model's forecast to guide development planning and day-to-day field management decisions.