Novel Method to Detect Fracture Closure Event Using Continuous Wavelet Transform

The Diagnostic Fracture Injection Test (DFIT) is widely used to get the fracture closure pressure, reservoir permeability, and reservoir pressure. Conventional methods for analyzing DFIT are based on the assumption of a vertical well but fail for horizontal wells drilled in ultra-low permeability reservoirs with potential multiple closures. There is still a significant debate about the rigorousness and validity of these techniques due to the complexity of the hydraulic fracture opening and closure process and assumptions of conventional fracture detection methods. In this study, a new signal processing approach was proposed by M.Y. Soliman, U. Ebru, F. Siddiqi, A.Rezaei, and I. Eltaleb (2019) and (2020) was extended to use the continuous wavelet transform to identify the closure time and pressure. The new method was applied to synthetic and actual field data. The synthetic data were produced using commercial fracture simulators based on fracture propagation and closure simulation principles with predefined fracture closure. To determine this closure instant, we decompose the pressure fall-off signal as the output of the fracture system into multiple levels with different frequencies using the continuous wavelet transform. This "short wavy" function is stretched or compressed and placed at many positions along the signal to be analyzed. The wavelet is then multiplied term-by-term by the signal, and each product yields a wavelet coefficient value. The signal energy is observed during the fracture closure process (pressure fall-off) and the fracture closure event is identified when the signal energy stabilizes to a minimum level. Because of the uncertainty of the real field fracture closure, a predefined simple synthetic fracture simulation with known fracture closure was used to validate the new methodology. The new continuous wavelet transform technique showed clear success without any prior assumptions or the need for additional reservoir data. The new methodology is also extended to actual field cases and showed the same success as conventional classical methods.