In our previous work we developed a series of dimensionless temperature drawdown curves describing heat extraction from a homogeneous EGS. Each curve represents a specific effective permeability (typical for EGS range was considered) and can be used to predict temperatures in the producer using given production rate, injection temperature, reservoir temperature, and well separation distance. This approach is applicable to both, 5-point and 3-point injection schemes and is being incorporated in a system dynamics model (integrated systems modeling tool for accessing multiple aspects of geothermal energy. In our current work we are extending this approach to heterogeneous conditions. Our first goal was to determine how the actual data, such as fracture apertures, spacing, and spatial distribution, can affect the reservoir behavior predicted with our homogeneous reservoir model. Our second goal was to understand the major factors controlling the differences in heterogeneous versus homogeneous reservoir predictions. Based on this understanding we wanted to develop a simple method for adjusting the dimensionless temperature drawdown curves to account for reservoir heterogeneity. To investigate the effect of heterogeneity we used the same numerical model as in the case of homogeneous reservoir, except the grid block size in vertical direction was smaller. The fractured reservoir was represented using fractured continuum model (FCM). At the grid block scale, FCM is an effective parameter model derived from knowledge of fracture network properties. The effective permeability and porosity were specified for each block as a function of fracture aperture and number of fractures in the grid block (some blocks did not have fractures) derived from geostatistical simulations. The maximum continuity was defined along the line between the injector and producer and the amount of continuity normal to that line is varied to change the anisotropy. The parameters for geostatistical simulations were derived from the literature review of the fracture properties common for granite (common medium for EGSs). Multiple permeability and porosity fields were generated for the different distributions of the fracture aperture and spacing. While the grid block permeability and porosity values were unique for each case, the mean permeability and porosity of the modeling domain remained the same. The results of the simulations were compared to the results for an equivalent homogeneous reservoir with permeability and porosity equal to the mean values used in heterogeneous simulations. This analysis enabled us to define the conditions in which the effects of heterogeneity are relatively small and the reservoir behavior can be sufficiently represented with an equivalent homogeneous reservoir. We also defined the conditions in which significant effects might be expected. Under these conditions reservoir can either behavior better (around 20% of all cases) or worse (the remaining 80% of the cases) than predicted with the equivalent homogeneous reservoir. We conducted an analysis of all the results associated with the cases in which the reservoir responses were significantly different from each other and from the equivalent homogeneous reservoir. We complemented this analysis with the analysis of the major flow paths from the injector to producer, time of travel along each flow path, and the flow path length. As a result, we identified a few major factors that control the reservoir response in heterogeneous conditions. The work is underway to incorporate these factors into the

Press the Back button in your browser, or search again.

Copyright 2012, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.