Advanced Forward Modeling Helps Planning and Interpreting NMR Logs

Abstract

Nuclear Magnetic Resonance (NMR) logging has gained its position in the industry as stand-alone interpretation or as an additional source of information to conventional log data. There are a wide variety of examples where NMR logging has disclosed secrets of the well that were not discovered by conventional logs, e.g. low resistivity pay zones identified by NMR logs but completely overlooked or viewed as not economically viable by conventional log interpretation. In some complex environments, NMR logging may even be the only alternative available to assess critical formation parameters. The interplay of rock and fluid properties, reservoir parameters, and NMR tool characteristics and settings shape the recorded NMR signals. Pre-job planning is critical for the success of the logging run especially now that NMR technology has gained more general acceptance and is being applied in complex environments. Integration of known parameters and estimated in situ properties may reduce the acquisition time, demonstrate the feasibility of a proposed logging run, and provide a handle to optimize the acquisition cycle. A priori knowledge will ultimately lead to better data and results in less time. We have built a state of the art, multi-fluid, forward model to simulate NMR responses under given conditions. The latest understanding of NMR physics is incorporated in the model, including free and restricted diffusion. The model is a versatile planning tool, linked to a very user-friendly graphical user interface, which offers easy input and great flexibility. Specifics of all currently available MRIL® tools and a broad range of rock and fluid properties are directly accessible through embedded database files. The user, however, has the flexibility to alter these values or to incorporate specific, regional knowledge. Polarization curves, composite transverse relaxation time (T2) spectra, and differential spectra are displayed. Computed fluid volumes, saturation levels, and relevant, mean T2 values are tabulated. The Planner is also being used more and more as an interpretation aid: geologists and petroleum engineers are testing their hypotheses by comparing computed NMR signals to actual logging data. Introduction Technology continues to drive our industry toward investigating reservoirs with grater complexity. Either the formation is more complex in it's lithologic variation or the setting calls for more complicated drilling methods, and often both are the case. Conventional logging methods are being used to provide quantitative information in mixed-lithology reservoirs, low-resistivity / low-contrast pay zones, low porosity / low permeability formations, medium-to-heavy oil reservoirs, when dealing with fresh formation water, etc. It has been well documented that these logging and interpretation methods have failed to delineate critical reservoir properties in these complex-logging environments. NMR logging may be the differentiating technique that can provide both a quantitative and qualitative assessment of these reservoirs. For NMR the logging environment is primarily defined by the formation being logged, it's temperature and pressure, and the drilling fluid that is being used. The industry offers a wide variety of drilling fluid products and each product can have an impact on the NMR data acquisition and/or interpretation. Thus today's NMR logging techniques must contend with a wide variety of environments and still provide the parameters needed to assess the reservoirs productivity. Thus, careful planning prior to data acquisition is critical for the success of the NMR logging run.