Enhanced Coalbed Methane Recovery with Respect to Physical Properties of Coal and Operational Parameters

Abstract

Abstract Modelling desorption-controlled reservoirs is a difficult matter owing to the complex nature of hydrocarbon transport processes. The complexity of enhanced coalbed methane (ECBM) recovery is not only caused by desorption-controlled behaviour of coalbed reservoirs, but also the interaction between coal media and the injected CO2/N2 gas mixture, which changes the coal properties with respect to injection time. In addition, the characterization of coal reservoirs and the determination of in situ coal properties related to the transport mechanism of methane are complicated, as there is a lack of standard procedures in the literature. In considering these difficulties, this study took the approach that relationships between rank and physical properties of coal be used for evaluation purposes. Parametric simulation studies with rank classification provided more representative results for coal reservoirs rather than univariate analysis. Besides coal rank, simulation cases were run for different reservoir types, well patterns, molar compositions of injected fluid and well types. The shrinkage/swelling process was taken into account by making use of the extended Palmer/Mansoori model. As a result, ECBM recovery was studied in field-scale with rank-dependent coal properties and different operational parameters in order to provide an overview. Introduction In the literature, there are parametric simulation studies(1–3) investigating the effect of each coal property on primary and enhanced coalbed methane recoveries. The trend has been generally to change a model parameter in a given range and to observe its impact on total methane recovery. In our previous study, however, a new approach was followed during preparation of input data for a commercial compositional simulator: Computer Modelling Group's GEM module(4, 5). Instead of using a real field or a hypothetical data set, rank-dependent coal properties in the literature were gathered to create a database. This database enabled us to acquire more realistic outputs from the simulator. In this paper, however, the rank-dependent database was used to simulate the effects of different coal properties and operational parameters on ECBM recovery. The following section gives information about these cases. Simulation Cases In this study, four different vertical well cases were run to observe the behaviour of our data set(5). In addition, horizontal well cases were simulated and compared to vertical well cases. All simulations started with the primary recovery of methane (CBM) which takes 10 years. After 10 years, the ECBM recovery technique was used for all cases except the methane-saturated dry coal case. The aim of 10 years primary recovery is to decrease the water saturation in the cleat system and to increase the relative permeability of the gas before the injection of the CO2 or CO2/N2 mixture. As there is initially no water in the fractures of methane-saturated dry coals, ECBM recovery is applied at the beginning. As for other modelling parameters, depth and thickness of the coal seam are taken as 915 m (3,000 ft) and 6.1 m (20 ft), respectively. The average geothermal gradient is about 2.5 to 3 °C/100 m. Initial reservoir temperature is calculated as 45 °C.