Geological Influential Factors on the SIS Well Productivity in Teviot-Brook Block, Moranbah Gas Field

Abstract

Moranbah coalbed methane (CBM) field locates in the Bowen basin, Australia. The main target formation is Moranbah Coal Measure (MCM) and surface in seam (SIS) wells are drilled for this middle rank coalbed methane filed development. However, great differences in fluid production and complicated geological influential factors of the SIS wells are the challenges to make efficient development. In this paper the drainage and production dynamic behaviors of over 33 SIS wells in Teviot-Brook block of Moranbah gas field are analyzed and the fou typical dynamic indexes are picked including peak gas production rate, typical gas production rate, peak water production rate and typical water production rate. The geological influential factors such as coal density, coal thickness, overlaying sandstone thickness, gas content (GC), permeability, dip, curvature, seismic attributes, etc. are studied based on the geological model and the key geological factors are identified. The principle component analysis (PCA) and neural network is utilized to identify the strong interaction which may exist between these factors and the typical gas dynamic indexes, and the production prediction model is built in Teviot-Brook block. The results show that the key geological influential factors are dip, overlaying sandstone thickness, multiplier of normalized GC and permeability, and curvature of the coal seam. The larger the dip and curvature of the coal seam, the higher the typical gas dynamic indexes; the overlying sandstone thickness has relative negative correlation with the typical gas production rate and positive correlation with the typical water production rate; the high value of the multiplier of normalized GC and permeability corresponds to the high typical gas production rate. Dim relationships exist between the coal thickness, coal density, fault distance and the typical dynamic indexes, which mean their more complex relationships. The production prediction model built using the inner relationships discovered by the neural network workflow delineates the gas production distribution and the high-yielding band from northeast to southwest in Teviot-Brook block. The geological factor indexes of high-yielding area are presented as follows: dip of the coal seam is greater than 4 degrees; the overlying sandstone thickness is less than 25 m; multiplier of normalized GC and permeability is larger than 0.13; and high curvature of the roof coal seam surface is preferred.