| Title | Impact of Fluvial Sedimentary Heterogeneities on Heat Transfer at a Geothermal Doublet Scale |
|---|---|
| Authors | Virginie HAMM, Simon LOPEZ |
| Year | 2012 |
| Conference | Stanford Geothermal Workshop |
| Keywords | fluvial heterogeneities, geothermal doublet, Paris Basin, Triassic |
| Abstract | Deep Triassic aquifers of the Paris basin are located in clastic series which are mainly composed of fluvial deposits. They lie between 2000 and 3000 m depth and their temperature can reach 120°C in some areas. There were few attempts (3) at geothermal exploitation in the early 80's but none of them was really successful (exploitation was performed during one year on one site but injection pump failure and well damage lead to abandonment). In the context of the renewal of geothermal exploration and development in the Paris basin several new projects are now considered. As a preliminary study of the specificities of Triassic aquifers, we investigated the impact of sedimentary heterogeneities of such environments on the hydrodynamic and thermal behaviour of a geothermal doublet (1 producer well and 1 injector well). The main objectives were i) to study the impact of these heterogeneities on the shape of the cold body developing around the injector and on the thermal breakthrough time, and ii) to explore a relationship between a heterogeneous model of permeability and a more simplistic model that could well describe heat transfer processes at the doublet scale. We used a 3D numerical block obtained with a process-based model of fluvial sedimentary deposits (Flumy model developed by Mines Paristech). This synthetic block is 5 km long, 4 km wide and 50 m thick. It is composed of 200 cells in x and z directions for 160 cells in y direction (6.400.000 cells). For numerical limitation, it was upscaled to 100 cells in x and z directions and 80 cells in y direction (800.000 cells). Doublet simulations were performed considering two cases: - the doublet is parallel to the main channels - the doublet is perpendicular to the main channels. Then the simulation results were compared to those obtained with a simple hydrogeological model with the heterogeneous model being replaced by a 2 layers model which cumulates sand bodies and clay bodies. We also studied the impact of using horizontal wells on thermal breakthrough in such heterogeneous environments. For each simulation a passive tracer was added to the cold injected water to compare the geometry of the cooled part of the reservoir to the injected fluid paths. |