Temperature Warmback Test Design and Interpretation in Presence of Long Injection History

Abstract

Abstract Fluids are often injected at different temperature than the target injection zone in the subsurface. Hence, temperature warmback is observed during well shut-in following injection of cooler fluid. Warmback analysis has been previously presented to determine the injected fluid distribution over the injection zone. However, when the warmback is preceded by a long injection history, warmback analysis can be challenging. The temperature/rate information for the long injection history may not be fully known and if known, it is computationally difficult to account for such large volume of data in the warmback analysis. More importantly, the long injection has likely perturbed the background temperature significantly through various thermophysical processes. Warmback analysis independent of the injection history should be enabled which is the purpose of this work. We introduce design considerations to conduct the warmback test so that the need for the prior injection history is eliminated. In addition, a simple graphical technique is introduced to enable analyzing the warmback data independent of the geothermal temperature which should be treated as unknown especially in presence of long prior injection. Warmback analysis must become independent of the initial temperature. The graphical method presented in this work requires plotting and line-fitting the observed warmback temperature versus its log-derivative with respect to the shut-in time ratio. The slope of the line-fitted data is inversely proportional to injectivity. Therefore, the graph provides the injection profile when plotted for various flow units (layers) of the laminated injection zone. The graphical method rigor is shown through its application to synthetic and field data.