Slug Testing in Multiple Coal Seams Intersected by a Single, Vertical Fracture

stimulated wells completed in commingled reservoirs, i.e., This paper examines tic transient pressure response from slug reservoirs in which the layers communicate only at the wellbore tests in wells completed in multiple, hya however, each layer is an isotropic, homogeneous, infinite, horizontal Conventional slug test type curves and analysis techniqucslta arc medium with a uniform thickness, permeability, and porosity, In based on single.layer, radiabflow models. Since coal seams arc addition, each layer is assumed to contain a slightly compressible rclativel~ thin compared to conventional natural gas reservoirs, fluid with viscosity and compressibility that are mdcpendcnt of economically . !ablc coalbed methane wells often must bc pressure, Finally, wc assume the initial pressures in all layers arc complctcd in multiple scams, In addition, the typical range of equal and single-phase flow exists at all times. Under these pcnneabilitkx observed in coal requires that wells be hydraulically conditions, the Laplace transform of the total, dimer,sionlcss fractured, especittll~ for the critical dcwatcring phase early in a pr:ssure response for the !aycrcdsystems,~is WCI1’S productive hfc. Thcrcforc, {hcse convcntiomd slug test analysis techniques arc not valid for either wtluating wells complctcd in multiple coal scams or assessing the post-fracture p’(u) = -----l-— i &j hj 1 potentialof hydraulicallyfracturedcoalbed methanewells, ,,, ,,, ,,, ,, .,,,. ,!!,!.,,,,,,,,,. .,. !(1) j., Ill P~j(”) Kawtki, et al,s examined the pressure behtwiorduring slug tests in WCIISintersecting infinite-conductivity, vertical frncxres, where u is tt Laphtcc spi~ccvariable, n is the numbs’ of layers, Similarly, Rushing, et td! investigated slug tcstin in wcus with andj is the layer index, ! finite.conductivity, vertical fractures, Both stu ies, however, were limited to single-layercases. Karsaki, et al,s dso studied the The dimensionless pressure, ~’j(u), in thejlh layer is computed shq test pressure response in unstimuhttedIaycredreservoirs, but with Cinco-Lcy,et aL’s7 model which considers a WC1l intcrscmd their study only consldcrcdhornogwteous,two-layer forrrmtions, by a fully-penetrating, tinitc.conductivity, vertical fracture, l’hc fracture is rnodcledas a homo ~eneous,uniform slab, Becausethe The purpose of this paper is to investigate the transicm pressure fructurc width is assumed to‘be much smaller than the fracxurc bchiwior during slug tests in wells completed in multiple, length and height, the model assuxs flow in the fracture is linear hydraulically fractured COJIscams, Wc dcvclopcd a model for and fluid influx at the fracture ti s is negligible, In addition, the model assumesfluid production 10 m the reservoir to the wellbore =-s and illustratio).sat cad of pa= occurs only through the fracture, Further, since the fracture