Laboratory Studies of Pilot Water Floods

Abstract

Published in Petroleum Transactions, AIME, Volume 219, 1960, pages 24–30. Abstract Procedures have been developed to study the performance characteristics of unconfined pilot water floods using scaled laboratory models. The effects of operating conditions on pilot performance for various well patterns have been investigated. The limitations of the laboratory model studies are outlined. Relationships between the oil recoveries of pilots and large-scale floods have been established. The applications of model studies to the prediction of large-scale water flood performance and pilot flood design are discussed. Introduction Pilot water floods are undertaken to evaluate operational procedures and to gain advance information about the performance of an extensive water flood. In general, however, the performance of a pilot flood is not directly indicative of what can be expected from a large-scale pattern water flood. This difference stems from the fact that the over-all flow configurations are not the same for the two systems. A water flood normally consists of an array of identical well patterns. In such an array, the perimeters of the well patterns are axes of symmetry and act as impermeable boundaries. Thus, an extensive, pattern water flood can be visualized as a repetition of "confined" floods. In contrast, a pilot water flood involving only one or a few well patterns is "unconfined". Here the well patterns are not balanced by other flood units; hence, the perimeter of the pilot area does not act as an effective boundary. Accordingly, only a portion of the fluids injected (and also of those originally contained within the pilot area) is captured by the producers of the pilot. The rest escapes into the surrounding reservoir. This usually causes the amounts of oil and water produced in the pilot to be different from those recovered by a producer in an extensive water flood. Moreover, the recovery observed in the pilot can be expected to be greatly influenced by operating conditions, i.e., by the bottom-hole pressures at the injection and producing wells. The above considerations show that full utilization of pilot test results and reliable prediction of large-scale water floods require a knowledge of the relationship between the recovery performance of confined and unconfined well patterns.