Prediction of the Fracture Closure Pressure from the Instantaneous Shut-InPressure ISIP for Unconventional Formations: Case Studies

摘要

Hydraulic fracturing is applied ubiquitously in unconventional hydrocarbon reservoir development toincrease the well productivity. To design a frac job, an injection falloff diagnostic test, such as minifractest, is conducted first to determine key formation properties and frac operational parameters, includingfracture closure pressure. In low permeability formations, the conventional pressure falloff analysis (i.e.G-function) is not practical to identify the fracture closure since it requires several days of well shut-in to collect enough pressure falloff data to reveal the fracture closure. In SPE-187495-PA, the authorsshow that it is possible to develop an empirical equation to predict the fracture closure pressure (Pc) frominstantaneous shut-in pressure (ISIP), the first falloff data point, by regressive analysis on datasets fromminifrac tests for conventional formations, including G-function estimated Pc, ISIP, petrophysical andmechanical properties. Since petrophysical and mechanical properties could be estimated from cores andwireline logs, the application of this equation requires a minifrac test to have a very short falloff period onlyto estimate ISIP. The objective of this work is to extend that equation for unconventional formations byintroducing appropriate deviation terms for tight-sand and shale formations, respectively, in order to reducethe discrepancy between predicted Pc and G-function estimated Pc. To this end, several datasets, each of which contain the same attributes, are collected from publicationsfor shale and tight-sand formations. Part of datasets are selected for developing respective deviation terms,for shale and tight-sand, to be added to the empirical equation, while the remaining datasets are used totest the respective new equation. Then a regression analysis is performed between Pc differences and theindividual petrophysical and mechanical properties for shale and tight-sand datasets separately. Eventually,two deviation terms have been derived and incorporated to the empirical equation, one for shale and anotherone for tight formations.