| Title | A New Numerical Model of Amatitlan Geothermal System, Guatemala |
|---|---|
| Authors | Enrique PORRAS and Julian XICARA |
| Year | 2020 |
| Conference | World Geothermal Congress |
| Keywords | Reservoir modelling, Dual-porosity, New generator type DELT, AUTOUGH2, TIM, Py-TOUGH, calcite scale |
| Abstract | The Amatitlan geothermal field has a high-temperature reservoir, with feedzone temperatures in most of the production wells reflecting boiling conditions. The existence of a shallow vapor-mobile two-phase region (i.e., a “steam capâ€) and a deeper extensive two-phase reservoir were confirmed by two wells (AMF7 and AMF8). Gas concentrations are variable but modest calcite scaling exists in AMF1RD. Temperature distribution suggests that the upflow of the Amatitlán geothermal system takes place beneath a significant area of the active Pacaya volcano. The revised conceptual model suggests that the outflow of the system flows laterally toward Lake Amatitlán. A three-dimensional numerical model consisting of 11,240 grid blocks was developed for reservoir simulations. Natural-state and production/injection simulations for near 12 years of exploitation were conducted using this model and AUTOUGH2 simulator, TIM pre-and post-processor and Py-TOUGH scripts. TIM and Py-TOUGH were essential for optimizing the model block structure, adding meteoric recharge and background heat into the model, assigning the topography to the top surface of the model and to represent Lake Amatitlán, all required for dealing with the complexity of the real geothermal system. Simulated temperatures and pressures distribution in natural state match field data measured prior exploitation. For production history matching a dual porosity model was developed and it reproduces current flow rates and enthalpy of produced fluid by the production wells. The model also reproduces pressure transient data of one monitoring well (AMF7). The effects of calcite scaling on production parameters of well AMF1RD is addressed in the model mainly by using the new generator type DELT, Yeh, A., et. al (2012). The model is capable now for predictions. As an example, a base case scenario was run; this base case scenario was divided into two different cases: Scenario one considers continue production as it is now for the next fifteen years without taking into account scale in the best production well AMF1RD. The second scenario is the same case as scenario one but it takes into account scale in wellbore and formation of AMF1RD. Results of simulated prediction scenarios are that by considering the scale issue during the simulation study, the simulated MW thermal are more realistic that the case that doesn’t consider scale. The difference in terms of MW thermal for both scenarios is about 2% lower for the most realistic case. Simulations also show a decline on MW thermal for both cases of about 0.90 MW thermal per year. |