| Title | Comparison of Modeling Results with Data Recorded During Field Stimulations at Utah FORGE Site |
|---|---|
| Authors | Pengju XING, Branko DAMJANAC, Zorica RADAKOVIC-GUZINA, Maurilio TORRES, Aleta FINNILA, Robert PODGORNEY, Joseph MOORE, John MCLENNAN |
| Year | 2023 |
| Conference | Stanford Geothermal Workshop |
| Keywords | numerical simulation, hydraulic stimulation, microseismicity, FORGE |
| Abstract | In April 2022, three stages of stimulation were carried out near the toe of well 16A(78)-32. During the field stimulations, pumping pressure, pumping rate, and microseismic events were recorded. Previously, before the actual field stimulations, predictions of stimulations were conducted using numerical models. The models used different discrete fracture network (DFN) strengths (from weak and initially permeable to strong and initially impermeable) and geometrical realizations. In this paper, modeling results were compared for three stimulation stages with the field data in three aspects: 1) injection pressure history, 2) spatial distribution of microseismic events, and 3) maximum event magnitude and b-values of microseismic events. After the comparison we find that all three stages appear to include combination of hydraulic fracturing and stimulation of DFN. DFN leakoff seems to dominate response in Stages 1 and 2, which is logical considering use of slick water. Stage 3, which is stimulated with xlink fluid, is dominated by hydraulic fracture propagation. The model overpredicts DFN stimulation for Stage 3, most likely because of approximation of power-law (xlink) fluid with Newtonian fluid (i.e., constant viscosity). Injection pressure histories for Stages 2 and 3 (cased completion with perforations) are not matched well because of complex evolving geometries and processes in well near field that are not included in this model. The b values of the microseismic events from the models are very close to those obtained from the field for all the three stages, which is 2.2 – 2.5. The models will be updated after the DFN is refined based on the field microseismicity. |