Search

To effectively control and utilise large amounts of gas emissions from underground coal mining, Australia and many other countries rely heavily on a series of vertical boreholes to capture the gas during mining. This gas capture not only reduces the greenhouse gas impact but also recovers a significant amount of energy. Vacuum pumps are connected to surface boreholes to drive gas flow from the goaf and overlying fractured strata. However, applying high suction pressure to the boreholes may cause more oxygen to enter the goaf from the longwall working face. This influx of oxygen can potentially react with residual coal in the goaf, accelerating coal self-heating and possibly resulting in endogenous fires. By analysing extensive goaf gas drainage data collected from Australian longwalls, researchers have established the normal trend of goaf gas concentration and flow rate as a result of intensive goaf gas drainage. This paper further developed goaf gas flow CFD models with the operating vertical boreholes, which were then calibrated using field borehole production data under certain operating conditions. The CFD model simulated the distribution of O 2 and CH 4 as well as the gas flow pathways in the goaf, considering the complication of goaf gas drainage. This study found that the O 2 concentration is higher on the goaf tailgate side under the influence of tailgate side boreholes compared to the maingate side. Furthermore, for the well-compacted goaf, a significant proportion of the ventilation air travels through 'high permeability flow channels' provided by the goaf edge on tailgate side, and then circulates back into the goaf boreholes. These CFD modelling results not only enable ventilation engineers to visualize the goaf gas flow patterns under the impact of multiple operating boreholes, but also to understand the impact on goaf atmosphere through sensitivity analysis of natural goaf characteristics, such as goaf permeability distributions and gas emission rates. [ABSTRACT FROM AUTHOR]

In gassy underground and spontaneous combustion prone coal mines, goaf management is a challenging task to reduce the risks of gas explosion and spontaneous combustion to acceptable levels. Goaf gas drainage via vertical surface goafholes is a primary control method to manage gas emissions in Australian coal mines. However, the operation of vertical surface goafholes maybe significantly impacted by insufficient understanding of goaf gas flow and coal self-heating behaviour, especially under different goaf gas drainage operating conditions. This paper focuses on analysing the field monitoring data collected from 161 goaf drainage holes at two coal mines in Australia. The data formed the basis to develop a conceptual model, which represents the changes of goaf atmosphere during the longwall retreat along with goaf drainage operations. The results indicate that the goaf drainage operating conditions may significantly affect goaf gas composition, which contributes to a different goaf atmosphere in comparison to existing models without goaf drainage. Within the studied longwall goaf, the 'oxidation zone' and 'inert zone' are identified with different carbon monoxide (CO) and oxygen (O 2) trends, and the 'oxidation zone' is conceptually further divided into the 'heat accumulation zone', 'elevated oxidation zone' and 'slow oxidation zone' based on the variation of gas composition and potential degree of oxidation of residual coal. The coal oxidation potential risk and corresponding goaf gas drainage strategies at these conceptual zones are also evaluated in this paper. The results of this study are expected to act as a guideline for future goaf gas drainage operations. • Gas composition analysis of goaf drainage production data was conducted. • A goaf environment conceptual model under various goaf gas drainage was proposed. • Coal oxidation risk at different goaf zones have been analysed. • Relevant goaf gas drainage plans have been proposed based on the conceptual model. [ABSTRACT FROM AUTHOR]

This paper evaluates the potential methane storage capacity of six clay-rich interburden rock samples from coal seam gas (CSG) wells in the Surat Basin, Australia. Clay minerals identified in all samples included kaolinite, illite, smectite, and illite-smectite mixed-layers. The total organic carbon concentrations in these interburden rocks ranged from 0.66–1.19 wt%, and thus these rocks can be classified as fair to good hydrocarbon source rocks. The effective porosity of the rocks determined from mercury intrusion porosimetry and helium pycnometry varied from 6.8% to 12.5%, and included volumes of micropores and mesopores. The adsorption isotherm results indicated that the average adsorption capacity of six interburden was 3.64 cm3/g, a value corresponding to approximately 20% that of Surat Basin coal. Based on the clay compositions and porosity of the samples, the permeability of these Surat interburden rocks is estimated to be <5 nano Darcy using Yang and Aplin's empirical correlation, which was too low for reliable measurement in our laboratory core flooding apparatus even with a differential pressure of 10 bar applied over a shortened 20 mm length core. However, after stimulation by electrohydraulic discharge (EHD) shockwaves the permeability of one of the interburden samples (S2) increased to 0.6 ± 0.11 mD due to development of fractures and new pores by the EHD stimulation. We characterised the development of the fractures after EHD shockwaves using x-ray computer tomography. The findings of this study suggest that dynamic shockwaves such as those generated by EHD have potential to increase permeability of soft and clay-rich interburden layers in CSG reservoirs and other layered reservoirs. This potentially opens these ultra-tight gas resources to exploitation and recovery. • A suite of coal seam interburden samples has been characterised and investigated for gas potential appraisal. • After EHD stimulation, the confining-constrained interburden discs were successfully cracked with fractures/voidages. • The shock pressure at both radial and axial orientations was recorded to reveal the philosophy of interburden breakage. • As a pioneering trial to stimulate interburden under stress-state, EHD was employed on stimulating clay-rich interburden. [ABSTRACT FROM AUTHOR]

Longwall coal mines in the Bowen Basin of Australia are highly productive and gassy, comprising many coal seams above the working seam. Predictive modelling of specific gas emission (SGE) using the conventional Winter model developed in Germany or other European based models often results in significant discrepancies from reality. To improve methane emission prediction and management, significant efforts on measuring gas release from different coal seams were made at three longwall mines by safely drilling core holes into the goaf. The vertical extent of the gas emission zone and the proportion of gas released from different coal seams were quantified. It was found that the Winter model significantly underestimates gas emissions for coal seams 75–150 m above the mining seam. This discrepancy is likely due to the intensive goaf drainage using large-diameter vertical wells in the high production Bowen Basin longwall mines. The Winter model also appears inadequate in estimating the height of the gas emission zone at different mines where the extraction thickness differs. Finally, a new empirical model for predicting the degree of gas emission from roof coal seams is developed for Bowen Basin mines and other mines with similar mining conditions. The measured post-mining gas content data and the newly developed model in this paper can provide guidance to the design of goaf drainage boreholes and capacity, as well as assessing coal mine methane and abandoned mine methane resources. [ABSTRACT FROM AUTHOR]

Significant increases in permeability of coal with continued production of coalbed methane (CBM) is a well-accepted phenomenon, particularly in the San Juan basin in the US and Surat basin in Australia. Modeling this increase is either based on the resulting increase in fracture porosity of coal or the associated changes in stresses as a result of the sorption-induced strain. This paper combines the experimental results of a study that measured sorption-induced coal matrix volumetric strain with depletion and a model proposed to estimate the associated changes in stress. The overall changes in stress, resulting from the combined effect of the poro-mechanical behavior and sorption-induced strain, were estimated by introducing a Biot-like coefficient. Plotting the stress path followed during depletion along with the failure envelope for the coal-type clearly showed that shear failure of coal is possible due to anisotropic loading resulting from a large reduction in the horizontal stresses. This would explain the large increases in permeability, typically observed in CBM operations. Finally, a permeability model was developed using the Biot-like coefficient, and assuming transversely isotropic behavior of coal. A comparison of the experimental and modeled permeability results showed that the model works well as long as coal does not fail. However, it demonstrates that a permeability model, incorporating failure of coal, is warranted for reliable prediction of permeability variation. [ABSTRACT FROM AUTHOR]

The sustainable in situ regeneration of microbial (biogenic) methane or microbially enhanced coal bed methane (‘MECoM’) is an emerging concept being investigated globally. Promising results and recommendations of a preliminary culture study of Walloon Subgroup co-produced coal seam gas (CSG) waters from the Surat Basin, Queensland established the presence of viable methanogens and suggested that in situ methanogenesis could be stimulated using physical and chemical reservoir treatments. This paper represents the culmination of a stepwise basin analysis project ultimately aimed at siting potential in situ bioreactor locations in the eastern Surat Basin. The integration of stratigraphically located data on the molecular and isotopic composition of desorbed gases, host coal properties and spatially associated waters through the core production zone has allowed the spatial variability and relative influence of hydro-geological factors on methanogenesis to be evaluated in detail for the first time. Higher gas contents and systematically enriched CH 4 and CO 2 carbon isotopic compositions in the stratigraphically central coal seams (oldest-youngest: in the upper Taroom Coal Measures, Tangalooma Sandstone lower Juandah Coal Measures) are best explained by increased rates of microbial CO 2 reduction and substrate depletion. There is a building case to trial MECoM in the central coal seams in a depleted/underperforming well in an area of high permeability. Integrated microbiological, chemical engineering, hydro-chemical and geological studies are ongoing to further enhance understanding of Walloon Subgroup CSG and the bioreactor potential of the Surat Basin. [ABSTRACT FROM AUTHOR]

In-situ stress pattern and subsurface fractures control the production of Coal Seam Gas (CSG) reservoirs and, hence, comprehensive analyses are required for better development of these reservoirs. The Surat Basin is a prolific CSG province with major economic impacts in eastern Australia. Recent studies have reported significant stress variations in this basin due to local geological structures. However, a systematic analysis of localised stress perturbations within the basin, and their impacts on reservoir permeability have not been presented thus far in the published literature. In this paper, we combine different sets of data including conventional log data, wellbore image logs, 2D seismic lines and interpreted permeability (from well test data) in order to characterise the role of geological structures in the subsurface fluid flow of the Walloon Coal Measures in eastern Surat Basin. Analysis of 7.3 km of borehole image log across 33 vertical wells revealed a regional ENE-WSW orientation of the maximum horizontal stress (S Hmax) in the study area. However, there are massive variability in the S Hmax orientation, spatially and with depth. Analysis of 2D seismic lines highlight the role of geological structures in stress perturbation, and the interpretation of natural fractures using image logs show that most of the open fractures are oriented sub-parallel to the acute angle of local S Hmax orientation. Interpreted well test data in the study wells demonstrates that most of the high permeability data points are located where in-situ stress rotation is observed. This indicates the influence of local S Hmax orientation responding to geological structures on the subsurface fluid flow in the eastern part of the Surat Basin. • Variability of S Hmax orientation laterally and with depth. • Small scale geological structures control the local in-situ stress pattern. • Open fractures are oriented sub-parallel to the acute angle of local S Hmax orientation. • Most of the high permeability data points are located where in-situ stress rotation is observed. [ABSTRACT FROM AUTHOR]

Density log data is one of the key physical attributes used for reservoir characterization by quantifying and qualifying lithological attributes in a wellbore. The density log is most often acquired by geophysical wireline logging techniques after drilling. However, wireline logging can be difficult to execute in highly deviated wells and alternative data acquisition such as logging-while-drilling (LWD) can be very costly. This paper describes the process of using instantaneous drilling attributes together with a machine learning algorithm, extreme gradient boosting (XGBoost), to generate a pseudo density log. The mean absolute error (MAE) and root mean square error (RMSE) are used as the evaluation metrics. Case studies are performed using data from six coalbed methane (or coal seam gas) wells in the Surat Basin, Australia. The inputs include drilling data [weight on bit (WOB), rotations per minute (RPM), torque, true vertical depth (TVD) and rate of penetration (ROP)] and LWD data (i.e., natural gamma ray and hole diameter). The MAE of pseudo density log for most wells is between 0.08 and 0.11 g/cc (except Well 4 is 0.16 g/cc), which results in an average error rate less than 5%. It is found that TVD, gamma ray, ROP, and hole diameter are four most important features. Further experiment shows that the model with these four important features has almost the same performance as the model with all features. The proposed machine learning methodology can assist petroleum engineers and geologists in reservoir characterization by generating pseudo density logs from ongoing LWD and drilling data in real time. It can potentially mitigate the need to run wireline logging tools after drilling or costly LWD techniques whilst drilling. • Pseudo density logs of six coal seam gas wells in Surat Basin, Australia are generated using a machine learning method (XGBoost). • Error rate of generated pseudo density logs for most wells is less than 5%. • True vertical depth (TVD), gamma ray, rate of penetration (ROP), and hole diameter are four important parameters. [ABSTRACT FROM AUTHOR]

Ultra-low permeability, inertinite-rich, poorly cleated, dehydrated, overpressured coal seams deeper than 9000 ft (2740 m) in the Cooper Basin of central Australia represent a fundamentally different play type to shallow and "deep" coal seam gas reservoirs. Proof-of-concept gas flow was achieved in 2007. Four Patchawarra Formation coal seams were subjected to low-proppant concentration slick-water hydraulic fracture stimulation within a dedicated vertical wellbore at a depth of 9500 ft (2900 m). Gas was produced for 81/ 2 years at a slowly increasing base flow rate, averaging 0.1 MMscfd (2.8 Mscmd). Experimental data were gathered for characterising dynamic reservoir behaviour. This included multiple extended pressure build-up tests. The authors have previously investigated the three largest of these using time-lapse pressure transient analysis. The published results reveal a dominant bilinear flow regime, associated with an isolated domain of increasing coal fabric permeability surrounding the hydraulic fracture. This paper builds upon the time-lapse pressure transient analysis study by specifically investigating conductivity of the hydraulic fracture, which the authors postulate to have also increased. The hypothesis is tested by applying time-lapse rate transient analysis to the two specially designed pressure drawdown tests to atmospheric pressure that immediately follow the first two pressure build-up tests of the time-lapse pressure transient analysis. Between the two pressure drawdown tests, spaced 586 days apart, the wellbore flowed gas continuously for 327 days, at an average flowing bottom-hole pressure of 580 psig (4.0 MPag). Hence, there was ample opportunity for hydraulic fracture conductivity to change between the tests. Both pressure drawdown tests were monitored for 24 h using high-resolution surface pressure gauges. Each was initiated from the same surface shut-in pressure of 2500 psig (17.2 MPag). The flow pressure data are initially used to construct "diagnostic plots" that clearly identify the hydraulic fracture linear flow regime, early in each test, immediately after the dissipation of wellbore storage. Hydraulic fracture properties are then extracted using "specialty plots" that display rate-normalised pseudo-pressure difference versus linear superposition time. Comparing the slopes of the two speciality plot trends indicates that the hydraulic fracture flow property b f k f ∅ f increased during the 327-day gas flow period by a factor of 4. This is supported by a 60% increase in the initial gas flow rate from "hydraulic fracture storage", from 7.5 to 12.0 MMscfd (212.4 to 340.0 Mscmd). Additionally, despite a significantly larger volume of "hydraulic fracture storage" gas being produced to surface during the second test, the duration of the hydraulic fracture linear flow regime is less than for the first. These observations are consistent with an increasingly more conductive hydraulic fracture over flowback time that allows compressed gas within it to discharge more rapidly. • A pilot experiment investigated dynamic behaviour of fracture stimulated Cooper Basin coal seams at 9500 ft (2900 m). • Time-lapse rate transient analysis was performed on two consecutive wellbore pressure drawdown tests. • The hydraulic fracture linear flow regime was identified by both tests. • An increase in hydraulic fracture conductivity was identified during the flow period between the two tests. [ABSTRACT FROM AUTHOR]

This paper presents the results of the surface hardness of coal, estimated with the Shore Scleroscope Rebound Hardness tester. The Shore hardness can be related to fracture toughness and tensile strength, which are important input parameters for fracture stimulation design of coal seam gas reservoirs. The Shore hardness experiments were performed on low rank coal specimens from an open cut mine from the Surat-Clarence-Moreton Basin, and a borehole from the Surat Basin, Australia. All measurements were conducted on polished coal surface confined within epoxy blocks following similar preparation required for coal petrographic analyses. The resulting indentations and microlithotypes were analysed under the microscope, then associated to the Shore hardness number. Three common microlithotypes were identified for the sub-bituminous specimens tested: vitrite, clarite and carbominerite. Results indicate the hardest microlithotype is clarite, followed by vitrite, and then carbominerite. Quantification of the micromechanical properties is possible using simple Shore hardness tests, and opportunities to improve fracture propagation may be further explored from this work. • The authors identified differences in shore hardness between microlithotypes when polished surfaces of sub-bituminous coal were indented non-randomly using a shore scleroscope rebound hardness. • The shore hardness of the clarite was greater than that of the vitrite and then carbominerite. • The vitrite band tends to show spread/propagation of the cracking and chipping (brittle microlithotype), although it is not always evident if the indenter reach a pre-existing crack which may have contributed to chipping observed on the bands. • The carbominerite is influenced by clay rich intervals (or porous occupied by clay and moisture) which can tolerate more strain (deformation) following an elastic-plastic behaviour ('ductile' microlithotype). • The rebounder of the indenter during a shore hardness measurement is softened on carbominerite when compared to a vigorous response of the rebounder on the clarite, while the vitrite is the easier to grind as it fractures and crack. [ABSTRACT FROM AUTHOR]

Identification of the extent of heat altered coals is important for coal mining and resource estimation because alteration directly affects key economic properties such as ash content, volatile matter, specific energy, sulphur, free swell index and beneficiation characteristics. The most reliable method to identify altered coal is through geochemical analysis of drill core samples; however this method is time consuming and costly. Cheaper alternative methods to identify altered coal is by macroscopic observation of non-core drill cuttings, but this is less reliable than core. Geophysical wireline log data is also used to identify altered coal, but with variable success. Over the past decades, Machine Learning methods have proven popular to automatically classify a variety of lithotypes from geophysical log data. However rarely has there been focus on predicting altered coal. In this paper we apply the most recent machine learning methods which include gradient boosted machines, random forests and artificial neural networks to automatically predict altered and non-altered lithotypes using geophysical log data. We use a massive data set comprising of 1230 samples from 263 boreholes from a highly intruded deposit in the Bowen Basin, Eastern Australia. To do this, we calculate each sample's distance from intrusion and predict their relative density from geophysical log inputs including gamma ray, caliper and compensated density. We then train our machine learning methods on 80% of the data to predict alteration class using the calculated distance to intrusion, predicted relative density, and geophysical log gamma ray, as primary inputs. We evaluated each of these machine learning methods on the remaining 20% of the data to determine the best performing model. Finally, using the best performing model we further split the altered and non-altered classification into lithotypes using a decision tree based on geological knowledge of the case study area. The results indicate that of the machine learning algorithms the random forest produced the best results with only 11 misclassifications across the entire data set of 1230 samples which represents <1% error. • Prediction of heat affected coals with a massive dataset and machine learning methods in the Bowen Basin. • Prediction of relative density and classification of altered and non-altered coals. • Lithology group definition based on a decision tree with <1% error. [ABSTRACT FROM AUTHOR]