Field application of capacitance-resistance models to identify potential location for infill drilling
A significant amount of bypassed oil often remains in a mature waterflooded reservoir because of non-uniform sweep. Infill drilling is one of the most attractive options to increasing oil recovery in consequence of its operational simplicity, low risk and promising results. Targeting proper infill location is a complex task and conventionally requires a comprehensive reservoir characterization program such as the streamline simulation (SLS). Achieving a good and reliable model, however, requires massive effort. This inspired the establishment of an alternative method, the Capacitance-Resistance Model (CRM), which is fast, cheap, yet robust. The CRM was applied to an oil field in Southeast Asia, leading to the identification of several key challenges and the emphasis of the input data examination. These challenges are field operations, existence of free gas, and unavailability of flowing bottomhole pressures, in which all of them cause the violation to the CRM assumption and will be addressed appropriately. In addition, this is the first time that the two-phase flow coupled CRM was used with field data. The key additional input required by this model is the reservoir pore volume associated with each producer, which is determined from matching each well’s historical water cut with Koval's equation. Nevertheless, dealing with field data is more complicated as the trend often did not follow the theory because of early water breakthrough in a thief zone or poorly managed waterflood. The results indicated that the CRM is able to give a good fit for both field and well levels. The quality of well by well matching seems to depend on the available number of data points of that well as all wells with low r-square values have very limited available data for matching. The gain results also reveal the good efficiency of the waterflood strategy as there are only 2 injectors having injection loss. It can also be geologically inferred from the gain that the field is anisotropic; there is no obvious preferential flow path in any specific direction. Moreover, field evidence such as the tracer tests, the RFT pressures and the wells' production history support the CRM results. The hypothesis made for the identification of the potential infill locations is that areas with low normalized gain, high oil saturation and high pore volume are attractive for new infill producers. This was successfully validated with the actual infill wells' performance of this field. The combination of maps consisting of the connectivity, the saturation, the thickness, the porosity and the permeability maps, are analyzed simultaneously to see whether the potential areas identified by them correspond with the performance of the infill wells. Generally, the integrated examination of these data is theoretically expected to help locate the bypassed oil and provide an insight to the reservoir characterization and the waterflood performance. However, it was observed with this set of field data that using more properties actually does not guarantee a more accurate result, especially when there are some properties that considerably mislead the interpretation. All in all, the combination of the relevant input parameters should increase the accuracy because they help each other to mitigate the errors caused from a single parameter. In other word, the potential area needs to have both poor reservoir continuity and good rock quality so that it is likely to yield a satisfying infill performance.