Interpretation of Hydraulic Fracturing Events by Analyzing the Energy of Rate and Pressure Signals

Abstract

AbstractHydraulic fracturing rate and pressure data are interrelated signals and can be subjected to signal processing techniques. The energy of a range of frequency band can be computed from the wavelet transform, which is a powerful technique in signal processing. The signal energies of the rate (cause) and pressure (effect) are used to detect the fracture events in time, such as height growth, screen-out, and hydraulic fracture-natural fracture interactions. The objective of this study is to relate the signal energies obtained from the wavelet analysis to the real physical phenomena of fracturing propagation events during injection. With this technique, the fluid injection during hydraulic fracturing process may be monitored in real-time for early diagnosis and to take preventive steps during the fracturing treatmentWavelet transform decomposes a signal into various levels of frequency components. The transform can be repeatedly applied to obtain a multiresolution analysis of the signal. Such repeated application of the wavelet transform yields the signal detail coefficients at each decomposition level, where each level represents a band of frequency. Using this approach, the pressure and rate signals from hydraulic fracturing were decomposed. Then, the pressure and rate signals energies at each frequency band were computed and compared to using energy density plots (EDP). Since wavelet transform preserves the time localization, all events on the EDP are obtained with respect to time. Finally, the energy distribution of the frequency bands for pressure and rate signals of different decomposition levels and using different wavelet types were studied with respect to time to distinguish between the rock-related and rate-related events in time. The moving reference point (MRP) technique is used to verify the results of our proposed approach.Several field cases were analyzed in this study to show the robustness of the proposed approach. The events identified using the proposed approach were in good agreement with the established techniques in the literatures such as Nolte-Smith and moving reference point (MRP). The identified events were related to physical events happening during fracture propagation, such as height growth, the opening of the natural fracture system, screen-out, and proppant entering the formation. The main advantage of the developed methodology is its ability to identify fracturing events more accurately and earlier in time than other techniques.The novelty of this research is detecting the fracturing events in time independently of the assumptions of fracture and well geometry. Ultimately, this technique helps to improve treatment designs and efficiency by analyzing fracture and formation behavior of the treatment and enhancing decision-making during execution, by providing real-time indications of fracture behavior.