Multi-lateral wells have been increasingly used in recent years by different industries including oil-and gas industry along with coal bed methane-and water production. The common purpose of these wells is to achieve a higher production rate per well. More and more sophisticated well patterns and geometries can be implemented in practice which calls for improved modelling techniques. Complicated well geometries and small lateral diameters require high resolution models in the vicinity of the wells. With structured finite difference grids this can only be achieved by unnecessary refinements even far away from the wellbores. However the model may still suffer from orientation problems if laterals do not coincide with the rows or columns of the rectangular mesh. In the present work, we applied unstructured grids to model multi-lateral wells and compared the results to structured models. We used the MODFLOW-USG code, which simulates groundwater flow using a generalized control volume finite-difference approach, allowing grids other than orthogonal structured grids to be applied. This offers a solution for orientation and resolution problems. The second part of the paper aims to optimize multi-lateral well geometry by evaluating the effect of length, angle and number of laterals.

Multilateral wells have been increasingly used in recent years by different industries such as the oil-and gas industry (incl. coal-bed methane (CBM)) and geothermal energy production. The common purpose of these wells in the oil industry and partially in the geothermal industry is to achieve a higher production rate per well without increasing the hydraulic gradient significantly. In geothermal systems the ultimate objective is to improve heat extraction performance by enhancing productivity but also by increasing the heat transfer area.Optimal design of multilateral wells has been the focus of numerous publications in CBM and unconventional oil-and gas production, but as a result of cross-fertilization and technology-transfer among different sectors in the energy industry in the last few years, the focus of research on multilateral wells has turned to enhanced geothermal systems (EGS).Recently, several papers have been published on the use of multilateral wells in deep geothermal reservoirs. Coaxial closed-loop geothermal systems with multilateral wells utilize only the increased heat transfer area due to the laterals while EGS's with multilateral injection and production wells also benefit from the hydrodynamic advantages of such well configurations. However, substantial emphasis isunderstandablyplaced on the effect of artificial and natural fracture networks of EGS reservoirs and less focus is devoted to the impact of vertical permeability anisotropy which can seriously influence the flow pattern of a given configuration. Similarly, in CBM-related studies, the effect of horizontal permeability anisotropy along the face cleats to the butt cleats on multilateral well-design was investigated and vertical permeability anisotropy was not considered.In the current work we present a generalized approach for evaluating the effects of vertical permeability anisotropy on multilateral well geometry by numerical hydrodynamic modelling in an idealized, homogeneous hydrogeological setting.Current paper proposes an alternative modelling approach for representing the non-horizontal branches of multilateral wells by introducing a thin layer for the laterals with the same inclination as the well branches. This approach is then is used to investigate the effect of anisotropy on the flow pattern and near-wellbore drawdown and hydraulic gradient in the case of different branch deviations. Results suggest that the benefits of highly deviated or horizontal laterals emerge when vertical anisotropy is high. Evaluation of branch deviation vs. anisotropy indicates that above approx. 60 • there is no significant benefit in increasing deviation which implies that very highly deviated or horizontal laterals might not necessarily pay off the associated technical challenges and extra costs.