Static pressure is one of the very important parameters for reservoir engineering, it gives us precious information about our reservoir, such as drive mechanisms, quantities of hydrocarbon in place, patterns, communication between wells, fluid behavior in the reservoir, as consequence, the measurement of this parameter must be conducted on periodical basis, to appropriately know the field and build a good model of reservoir.

The advantage of this study can complete other studies that concentrate only on the oil production rate forecasting like Data Driven Production Forecasting Using Machine Learning [1], Production Forecasting in Conventional Oil Reservoirs Using Deep Learning [2], Machine Learning Prediction Versus Decline Curve Prediction: A Niger Delta Case Study [3], Decline Curve Analysis for Production Forecasting Based on Machine Learning [4] ……, in addition of static pressure evolution of wells.

For instance, we can optimize through this study a number of conducted tests to measure static pressure which will minimize operating costs and the probability of accidents occurring the operations, also reduce the shutdown time of wells for completion purpose of such measurement, in addition to the possibility of using this model for other analogue wells that do not have enough pressure measurement, without the need for time and extensive study. Besides, multivariate polynomial regression machine learning algorithm has been developed in this study to predict the evolution of static pressure for existing oil wells.

You can access this article if you purchase or spend a download.