Production Data Analysis in Eagle Ford Shale Gas Reservoir

摘要

Eagle Ford shale in South Texas is one of the recent shale play in US which the development began in late 2008. So far many horizontal wells have been drilled and put into production using hydraulic fracturing. Production behaviour in shale gas reservoirs unlike conventional reservoirs is different in various plays and there are no published reports for production data analysis in Eagle Ford shale. We have used the linear dual porosity type curve analysis technology to analyse the production behaviour and to estimate the essential parameters for this reservoir. This type curve was constructed based on transient production rate with constant well pressure in a closed boundary of stimulated reservoir volume (SRV) with double porosity approach. In order to analyse the early production data we used Bello’s and Nobakht’s approach to account for apparent skin. In this study, three flow regimes were identified consisting of 1- bilinear flow; 2- matrix linear flow; and 3- boundary dominated flow. For the analysis of early flow regime, two possibilities of transition flow and apparent skin have been considered. First, the fracture permeability was estimated to be around 820 nano Darcy based on transition flow analysis. Second, the matrix permeability was estimated to be either 181 or 255 nano Darcy based on two different approaches in matrix linear flow regime. Furthermore, original gas in place (OGIP) and SRV were estimated from the boundary dominated flow regime. To validate the estimated matrix permeability, a single porosity numerical model with high permeability transverse fractures was built to match the production history. The permeability from simulation was in a good agreement with type curve analysis. Production forecasting has also been carried out using different adsorption isotherms. The results showed that the effect of desorption depends on the reservoir pressure and the shape of adsorption isotherm curve. In early time of production, desorption is usually not effective, however, for long-term production forecasting, it is necessary to account for this phenomenon by providing an accurate isotherm.