| Abstract |
Most geothermal reservoirs are observed to be fracture dominated and are usually modeled with dual porosity. Well testing is a tool for better understanding and describing these types of reservoirs. Properties of naturally fractured reservoirs are critical parameters to acquire to enable accurate dual porosity reservoir modeling. Nonetheless, up to now well test analysis and reservoir simulation for dual porosity model are not well integrated. The aim of this study is to correctly characterize the reservoir properties, e.g., the permeability, storativity and interporosity flow. A simple synthetic model of a reservoir is built to facilitate the understanding of a well test. It employs several assumptions as cited by Warren and Root, 1963. A build-up test was simulated for 3 months to obtain adequate data for analysis. Pressure and flow rate as a function of time resulting from the build-up test are used in a well test simulator. The evaluation is done by matching the curve and the resulted key parameters can be interpreted in terms of reservoir properties. A sensitivity analysis has been used to investigate the effects of reservoir properties on well test analysis for a dual porosity system. Some of reservoir properties obtained are close to the given values of the synthetic model, such as fracture permeability, matrix permeability and fracture spacing. Volume fraction of fracture reflects a difference, which might be caused by the shape factor used. A case study was also carried out to validate the methodology of this study by comparing measured PT data and simulation result. The result indicate a slight different between measured data and simulation result. Well test analysis has proven very important for reservoir simulation, in order to get better results and accurate predictions of reservoir behaviour. This study helps to simplify the reservoir simulation process because reservoir properties to be used for reservoir simulation can be obtained from well test analysis. Thus well test results will provide a useful basis for reservoir simulation. The more accurate the reservoir properties, the better will be history matching and future predictions. |