Fractured water injectors are quite often the rule rather than the exception in many of the water floods operated by Shell world-wide. Fractures are often caused by thermal effects or created deliberately usually as a consequence of trying to reduce costs through re injection of produced water. In any case, it becomes important to have reliable tools available which are capable of predicting the lateral and vertical extent of these fractures, to assess their impact on well injectivity water flood sweep efficiency and the likelihood that unwanted communication is established with adjacent formations. There has been a considerable amount of development of predictive tools in this general area, both within and outside Shell. Within Shell, for screening type application, only fracture growth models in water injection wells were available assuming complete fracture containment (i.e. rectangular fractures). Also, these models did not account for the effects of damage on fracture growth. For this reason we have developed a new screening tool assuming penny-shaped fractures. As in pervious work, we have included poro- and thermoelastic backstress effects, together with a simple model for describing the effect of impurities in the injection water on fracture growth. We believe that the assumption of a penny-shaped fracture offers a more realistic description of the fracture propagation mechanism, particularly in thick formations or in thinner formations at early times. The new fracture propagation model is entirely based on analytical solutions to the equations of fluid/heat flow and to those of poro- and thermoelasticity. The model assists in the integration of surface and subsurface technologies by allowing engineers to quickly estimate the effect of changes in injection water quality on well injectivity via the effect on fracture growth. The new model also provides a useful benchmark when considering even more complex situations, such as the inclusion of vertical stress gradients and contrasts in stress. The paper also demonstrates the application of this screening tool to several of Shell's water floods in the North Sea and the Middle East whilst briefly describing the theoretical development behind the model.
Shell operated EP companies currently produce ca 0.5 million m3/d oil and a slightly larger quantity of water. The volume of water produced from many maturing fields is increasing significantly with time and, in the absence of remedial measures, will soon reach 1 million m3/d by the year 2000. Increasing water production means increasing operating costs to maintain existing facilities and to dispose of the water in an environmentally benign manner. Therefore, efficient water management will be a key element in the drive to contain costs. High pressure water injection under fracturing conditions using minimal water treatment may be the most cost effective water disposal scheme. In such a scheme, subsurface constraints may arise from considerations of the impact of injection on the recovery process, in which case fracture size is important, or from environmental constraints where the need to limit water injection to a given horizon may be essential.
Within Shell, Koning and Dikken developed models for propagating waterflood-induced fractures which included poro- and thermoelastic backstresses but made the restrictive assumption of a constant fracture height. Choate, on the other hand, developed a 3D fracture growth model primarily aimed at the prediction of fracture size during stimulation treatments of producing wells. This model, does, however, not include poro- and thermoelastic backstress. The present model, which is meant to be a screening tool for predicting the growth of water flood induced fractures, should be seen as an attempt to bridge the gap between the two classes of models described above. P. 179
