One of the most important objectives of transient well-test interpretation is to recognize the well-reservoir system (identification) that is being tested, before estimating its parameters such as permeability, skin, and reservoir pressure. This paper presents a procedure for system identification and parameter estimation for single layered reservoirs using downhole pressure and flow rate measurements. However, the procedure can also be applied to commingled reservoirs if each zone is tested in accordance with the multilayer testing technique. The procedure consists of three steps. The first is a system and flow regimes identification (diagnostic) from deconvolution, and specific convolutions and their derivatives. The second step is to obtain all possible parameters and to refine the model further from flow parameters and to refine the model further from flow regime analysis and type curve matching using the rate normalized pressure and its derivatives with a few selected models. Any parameter estimated from the flow regime analyses is used as initial guesses for type curve matching. The final step is the verification that allows an automatic estimation of parameters using the most probable model. At the final step, the selected model probable model. At the final step, the selected model should satisfy all observed transient measurements and the past production history as well as geological and well log information.
The procedure for transient well test interpretation presented in this paper demonstrates the advantages of presented in this paper demonstrates the advantages of using downhole flow rate and pressure measurements. In addition, the paper explores the use of the Gladfelter deconvolution of pressure and flow rate data for model identification. The convolution type curves (CTC) are presented. The interpretation procedure was presented. The interpretation procedure was successfully applied to a field well test example.
During a transient well test, if the change in the sandface flow rate together with the corresponding pressure response are measured, in theory it is then possible to identify the system topology and to estimate the geometrical and flow parameters. During the last forty years a great deal of effort has been concentrated on measuring the reservoir response to a rate pulse as a pressure signal at the sandface. These efforts have been pressure signal at the sandface. These efforts have been successful and it is now possible to measure sandface pressure with an accuracy of one tenth of a pound per pressure with an accuracy of one tenth of a pound per square inch. On the other hand, the sandface flow rate input (pulse) to the system has not been measured continuously, but the surface flow rate is usually regulated by a choke or other mechanical system. In some cases, it has been measured sporadically as a function of time in the gathering tank after the separation of gas from crude oil. These types of measurements of flow rate have three main drawbacks: first, the fluid seen by the pressure sensor is quite different than what is measured pressure sensor is quite different than what is measured at the wellhead or in the tank, second, there is a considerable wellbore volume between the pressure sensor and the wellhead where the flow rate is measured, and third, both pressure and rate measurements do not belong to the same time span.