High Performance Software Computing Enables Accurate Downhole Pressure Predications for Gas Wells

Abstract

A reservoir's static bottomhole pressure is an integral component of many reservoir evaluation disciplines. The static bottomhole pressure is normally acquired through gauge measurements; however, this method has disadvantages such as cost and mechanical risk. Accordingly, the ability to accurately estimate the static bottomhole pressure would provide a cost-effective and safe alternative to well intervention. In this work, a new cloud computing method is introduced to predict the static bottomhole pressure of a natural gas well. The method reaps the benefits of available IR 4.0 technologies, namely multi-layered high performance software computing. The utilization of an advanced software codes enabled accurate and timely prediction of gas wells’ bottomhole pressures. This method differs from existing methods by utilizing the apparent molecular weight profiling concept. Based on the inputs of pressure and temperature gradient data, an iterative calculation scheme is applied to produce a well-specific molecular weight profile. This profile is used along with a modified form of the equation of state to perform top node pressure calculations and ultimately predict the static bottomhole pressure for gas wells. The new calculation method was applied on two calculation modes: calibration mode and time lapse mode. In the calibration mode, the static bottomhole pressure is predicted on the same gradient survey used to generate the apparent molecular weight profile. On the other hand, the time lapse calculation mode predicts the static bottomhole pressure after a period of time has elapsed from the gradient survey used to build the molecular weight profile. The top node method was tested rigorously, and the prediction results were found accurate with low error percentages.