Diagnosis of Horizontal Injectors with Distributed Temperature Sensors

Abstract

Abstract We present an interpretation methodology for analysis of temperature profiles obtained in horizontal injectors with fiber optic distributed temperature sensors. We consider injection as a non-isothermal process, due to the contrast in the temperature of injected fluid and reservoir fluid. We examine the particular case of water injection in an oil-bearing reservoir and analyze a field case with a notable temperature contrast downhole (222 °F reservoir, 100 °F water). We examine the behavior of the wellbore temperature profile during injection, shut-in, and post-shut-in periods, for both short-term and long-term injection (days to months). We present a methodology for estimation of the injectivity profile along the length of the well during short-term injection, and discuss issues associated with interpretation of long-term data. The proposed methodology can be applied to injection testing of producing wells. Introduction The use of horizontal wells as injectors is now relatively common in oilfield operations, due to the potential for injection of large volumes of fluid per well and the possibility to reduce the total number of wells required for injection. Flood efficiency, however, depends greatly on the degree of uniformity of sweep. To enhance sweep it is usually desirable for the injection profile in wells to be as uniform as possible. In horizontal injectors, this is a particular concern, because the contact with the formation can span one to several kilometers, and the well can traverse a sequence of geological facies, including intervals of high hydraulic conductivity or fractured zones. Currently, to assess the injection profile in a horizontal well, it is required to conduct a downhole survey with the aid of ‘production logging’ sensors and in certain cases conveyance with coiled tubing may also be required. We present in this paper a method for determination of injection profile in horizontal wells that does not require physical intervention in the well. The method, however, does require deployment of distributed temperature sensors in the well during the completion stage, which is possible with fiber optic lines. These lines provide a temperature trace or ‘log’ with a spatial resolution of about 1 meter, temperature resolution of about 0.2 °F, and an acquisition frequency that can be virtually continuous (every 7 seconds), although many applications can be adequately addressed with periodic surveys (e.g., monthly).1 Temperature logging is not new in the oil industry and the principles for interpretation of temperature logs have been elucidated at least as early as 1950s.2–7 Nevertheless, there are two reasons why this body of work should be revisited. First, because the classical methods are applicable to vertical well geometries; and second because these methods account for the dimension of time in a relatively limited sense. In this work, we will examine to what extent concepts developed for interpretation of temperature logs in vertical wells are applicable to horizontal wells, and present a specific procedure for analysis of horizontal well temperature profiles to infer the injectivity profile. We also make a preliminary investigation of the effect of injection time on the wellbore temperature behavior and assess how this impacts the validity of the interpretation technique for injectivity profiling. We conduct this work numerically, using a reservoir simulator that has a wellbore modeling facility for fluid flow and heat flow calculations.8–9 A concurrent work that will be published separately, investigates this problem analytically.10 Numerical Model The simulation model is 110 ft thick with an areal extent of 7000 ft in X direction and 4100 ft in Y direction. To simulate the effect of cap and base rock, two nonreservoir layers of 50 and 100 ft thick are added to the top and bottom of the simulation model. The temperature of the reservoir is initialized according to the geothermal gradient of 0.02 °F/ft with the temperature at well horizon of 222 °F. The reservoir is initially filled with oil. The composition of the reservoir fluid is 40% C1 and 60% C7~. Crude composition is specified because the simulator which contains the wellbore temperature modeling facility is compositional. Therefore, a simple composition is specified to be able to use this facility (i.e., the model is black-oil). The temperature of the injected water is 100 °F. The main parameters of the simulation model are summarized in Table 1.