CASTEM—A New Automated Parameter-Estimation Algorithm for Single-Well Tracer Tests

Abstract

Summary The single-well chemical-tracer (SWCT) method has been used in more than 140 tests since 1968 to determine residual oil saturation (ROS) after waterflooding. Interpretation of the test involves matching tracer data from the field by use of computer simulation programs. As the complexity of the simulator model increases, the number of trial runs needed to fit field data satisfactorily increases rapidly. The conventional fitting procedure can thus become very cumbersome and can involve prohibitive computer costs. A new method has been devised to model tests in heterogeneous formations. These rigorous simulators have been replaced by simple response functions capable of being used in a standard parameter-estimation routine. Matching the tracer data now involves function evaluations rather than full simulator runs, resulting in a large reduction in computing time. The algorithm developed here includes many frequently used tracer test models. The new method has been used to explain both simulated and actual field data from tracer tests conducted in multilayer reservoirs. The choice of an appropriate model to explain a specific set of field data may be a qualitative decision, especially in the absence of other supportive physical evidence. However, the variation in the effective ROS predicted by the different models was found to be quite minimal and within the limits of accuracy of the test.