Sensitivity analysis of methane hydrate bearing Class 3 reservoirs during thermal injection

Sensitivity analysis is required for the different geological settings of methane hydrate-bearing reservoirs in order to understand the effects of reservoir parameters (such as permeability, porosity, sediment grain density) as well as the role of production parameters (such as injection rate, injected fluid salinity) during methane production. In this study, a sensitivity analysis was conducted using the one-factor-at-a-time (OFAT) approach of thermal stimulation in a Class 3 reservoir. A 2-D cylindrical grid was chosen such that mesh convergence was reached in both radial and vertical directions. Upon analysis, injection flow rate was obtained to be the most crucial parameter since the process of dissociation was driven thermally and this parameter dictated the net thermal energy supplied to the reservoir. The second most important parameter was identified as permeability in the range of 10mD to 30mD. Incidentally, permeability in the range of 50mD to 100mD was the least crucial parameter. This was because increase of permeability in the lower ranges significantly increased the dissociated methane volume due to improved thermal energy distribution due to improved fluid flow. However, at higher permeability, further improvements in fluid flow did not lead to increased dissociation since the thermal energy was limited by the injection flow rate. Thus, artificially increasing permeability in the hydrate bearing reservoir is crucial up to an optimal permeability value, beyond which the dissociated volume would be limited by the net thermal energy available. Multiple production scenarios were also studied using the huff and puff method with varying injection duration, soaking times and production rates w.r.t their recovery efficiency. The scenario which did not include any soaking stage outperformed all the other scenarios. Also, prolonging the duration of the injection duration exhibited a decline in gas produced. The energy efficiency was calculated for the most optimal scenario, but even that was found to be economically infeasible.