One of the keys to successful and environmentally responsible well stimulation programs in coal seam gas development is to establish consistent procedures for the safeguarding, planning and executing activities across multiple wells. The aim of this paper is to show how a novel application of petrophysical program scripting can be used to make the stimulation process more efficient, consistent and compliant across assets with varying requirements. A macro embedded in petrophysical evaluation software applies a series of rules to rank coals by thickness, allocate a series of perforations and stimulation type based upon coal rank and spacing and then produces actionable treatment schedules which are seamlessly implemented in well stimulation operations at well sites. To do this, the macro grades all coals within the well by thickness based upon a cut-off on the density log, with the thickest coal being graded highest. The macro then identifies the top ranked coal and places perforations based on user defined logic, geological information from offset wells, permeability attributes of the target coal layer(s), depth and vertical separation between adjoining coal targets. Based on the stimulation type assigned, a stimulation schedule is generated that includes estimates of fluid volumes, proppant volumes, injection rates, proppant ramp type and stipulates flush conditions (over-flush or under-flush). Coals thicker than a maximum perforation size are perforated in an upper, middle and lower configuration. Most coals are thinner than the maximum allowable perforation interval and so the macro looks up and down the borehole to include thinner coals within a potential perforation window. The system then generates the stimulation schedule as described above. The macro continues to allocate perforations and stimulation schedules for each validated coal interval and sequentially tries to maximise the total target coal interval along the wellbore. Certain environmental constraints are included in the macro logic to maintain local and regional commitments. For example, coal zones in proximity of permeable non-coal layers i.e. interburden are automatically excluded from stimulation. Multiple advantages of this system have been realised including, a) effective QA/QC as outputs can be directly plotted against the well logs giving the user a quick and easy visual check b) actionable instructions that site based teams can execute including exact perforation depths and stimulation schedules c) provide realistic materials and costs estimates that ensure efficient planning and logistics, d) monitor and document any variations between allocated schedule versus actual execution, e) provide estimate of expected net coal connectivity at a well, development package and asset level which feeds into production and recovery forecasts, f) plan future optimisation studies or pilots and g) most importantly offers a consistent, efficient and compliant framework that can be applied across multiple assets, engineering teams and service providers. This paper focuses on capabilities and advantages of using a macro to automate stimulation design allocation for CSG multi-well (>100 wells) assets. Details of individual stimulation designs for Walloons Coal measures are mentioned in other publications (Kirk-Burnnand et al., 2015 and Flottmann et al. 2018) and hence not covered here.

The Walloons coal measures located in Surat Basin (eastern Australia) is a well-known coal seam gas play that has been under production for several years. The well completion in this play is primarily driven by coal permeability which varies from 1 Darcy or more in regions with significant natural fractures to less than 1md in areas with underdeveloped cleat networks. For an economic development of the latter, fracturing treatment designs that effectively stimulate numerous and often thin coals seams, and enhance inter-seam connectivity, are a clear choice. Fracture stimulation of Surat basin coals however has its own challenges given their unique geologic and geomechanical features that include (a) low net to gross ratio of ~0.1 in nearly 300 m (984.3 ft) of gross interval, (b) on average 60 seams per well ranging from 0.4 m to 3 m in thickness, (c) non-gas bearing and reactive interburden, and (d) stress regimes that vary as a function of depth. To address these challenges, low rate, low viscosity, and high proppant concentration coiled tubing (CT) conveyed pinpoint stimulation methods were introduced basin-wide after successful technology pilots in 2015 (Pandey and Flottmann 2015). This novel stimulation technique led to noticeable improvements in the well performance, but also highlighted the areas that could be improved – especially stage spacing and standoff, perforation strategy, and number of stages, all aimed at maximizing coal coverage during well stimulation. This paper summarizes the findings from a 6-well multi-stage stimulation pilot aimed at studying fracture geometries to improve standoff efficiency and maximizing coal connectivity amongst various coal seams of Walloons coal package. In the design matrix that targeted shallow (300 to 600 m) gas-bearing coal seams, the stimulation treatments varied in volume, injection rate, proppant concentration, fluid type, perforation spacing, and standoff between adjacent stages. Treatment designs were simulated using a field-data calibrated, log-based stress model. After necessary adjustments in the field, the treatments were pumped down the CT at injection rates ranging from 12 to 16 bbl/min (0.032 to 0.042 m3/s). Post-stimulation modeling and history-matching using numerical simulators showed the dependence of fracture growth not only on pumping parameters, but also on depth. Shallower stages showed a strong propensity of limited growth which was corroborated by additional field measurements and previous work in the field (Kirk-Burnnand et al. 2015). These and other such observations led to revision of early guidelines on standoff and was considered a major step that now enabled a cost-effective inclusion of additional coal seams in the stimulation program. The learnings from the pilot study were implemented on development wells and can potentially also serve as a template for similar pinpoint completions worldwide.

Bulk permeability of coal is a critical parameter in coalbed methane (CBM) or coal seam gas (CSG) well completion designs and field development planning. The estimation of permeability can be made by well testing either during drilling or production; however, well tests are costly, time sensitive and resource-intensive. Therefore, field-wide estimates are often dependent on production data history-matching, which has a high degree of uncertainty. In this paper, we present a new attempt to apply machine learning approach to estimate coal permeability using drilling data. We first extract important parameters from well test analyses, which are obtained using a packer element testing (PET) tool from four wells in the Surat Basin, Australia. Then drilling data from the wells are processed and fed into different artificial neural networks (ANNs), which include multi-layer perceptrons (MLP), convolutional neural network (CNN), and recurrent neural network (RNN). Two types of models are constructed: regression model with permeability values, and classification model with permeability class intervals (i.e., low, medium, and high permeability values). The evaluation metrics include R2 for regression models and confusion matrix for classification models. Results show that the drilling mud losses are generally higher in coal layers and lower in interburden formations. The predicted medium and high coal permeabilities from MLP and CNN are in good agreement with measured permeability values from PET data. For the classification task, the CNN achieved an overall accuracy of 99%. Thus, an improved coal permeability map with a higher resolution and less calibration against PET data can be developed quickly to aid production data history matching. The developed machine learning model demonstrated a potential to be applied to new wells to predict the coal permeability for the Surat Basin as well as other CSG appraisal projects. This model can allow more rapid optimisation of well spacing, improved downhole pump design, more targeted well stimulation, and overall project economics.