Abstract
Mature-fields production optimisation faces many challenges, among which geoscience-related problems are of paramount importance. The Welfare method developed by TOTAL addresses many of these problems, as it is aimed at characterising reservoir behaviour and detecting insufficiently drained areas, as well as proposing and testing re-engineering scenarios. The method uses an in-house developed software that is based on mono-variant and multi-variant statistical analyses of mainly production data, as well as geological, reservoir, and fluids data. It includes an iterative integration of all, quantitative or qualitative, geoscience and production data with the statistical computation results.
Each well production history is summarized by selected characteristic values, called Production Indicators, that are calculated and mapped all over the reservoir area. In this way, field compartments and sectors with different production characteristics clearly appear and are interpreted with the help of static and fluids data.
Previously unknown features often appear, such as production-impacting fractures (sealing surfaces, fracture corridors, axes of water or gas breakthroughs, etc.), areas with increased vertical reservoir connection, spatial distribution of impacting heterogeneity, or field sectors of different value. The uniformity, or to the contrary, a large variation range, of well production behaviors throughout the reservoir area indicates the degree of lateral continuity that characterizes reservoir properties. The knowledge of this, and related interpretations, led to a reservoir diagnosis, and to a qualification of the degree of predictability of production behavior of future infill wells. These results are subsequently used to determine the location of under-produced areas, and a new scenario of future field re-engineering can then be tested and future production calculated. An example of fluvial-deposited silici-clastic reservoirs shows why the field had to be divided into separate areas for water injection. A second field, made up of vertically repeated deltaic sandstone reservoirs, could be split into significant areas with a clearly distinct dynamic. Another example concerns a platform carbonate reservoir that developed on a paleo-high. Calculations with Welfare proved, in a very fast way, an increased vertical connectivity near the top part of the structure that is fully compatible with the sedimentary and diagenetic settings. This knowledge and the distribution of, for instance, produced water salinity led us to proposing a modification of the water injection pattern and additional development drilling that proved very productive.
The experience acquired over a few tens of fields demonstrates that most of the fields can be processed with the Welfare method and software. Fields with long production history and many wells (say more than 40 years of production and more than 1500 wells) can be easily processed, as demonstrated by one of our examples. Our experience was gained from reservoirs covering most of the silici-clastic and carbonate depositional environments, and a variety of structural settings and fluids characteristics. Welfare is a key tool to IOR as it allows spotting un-drained hydrocarbons and thus increasing the quantity of recoverable reserves, and as it facilitates the optimisation of field re-engineering.