A Comparison of Tidal Signal Extraction and Bourdet Smoothening for Removal of Tidal Effect Induced Artifacts in Pressure Transient Analysis

Abstract

The results of pressure transient analysis (PTA) are very important in reservoir characterization; however, this analysis can be affected by some non-reservoir behavior such as gas breakthrough, phase segregation in the wellbore, tidal effects, all of which can perturb the result accuracy. When data is acquired for PTA offshore, it can contain tidal effect, causing noise which can lead to misinterpretation when the test is analyzed, hence its impact should be accounted for in the analysis. Tides are experienced as the rise and fall of sea levels due to the variation in the earth's gravitational potential exerted by the moon and the sun, and the rotation of the Earth. Tidal signals have been observed to mask late time response for pressure build up tests and will significantly hinder correct interpretation of reservoir boundaries if left unaddressed. The effects of tidal pressure signals on the pressure derivative of pressure build-up tests are studied with the aim of comprehensively exploring the deviation from expected responses given known reservoir boundary conditions. Subsequently a refined method for pure tidal component removal from pressure derivative data is presented and compared to simpler Bourdet smoothening (L) and filtration of data points used in evaluation. This work focused on an efficient method to analyze data containing tidal effects. The Bourdet derivative and log cycle filtration was effective in removing tidal signal effects on late time boundary identification with the drawback being having multiple possible interpretations of the IARF. Extracting the tidal signal gave a more defined IARF period and late time boundary effect period with only minor oscillations in the late time but the rigor of extracting the tidal signal without sufficient regional tidal information may prove to major hindrance to this process.