Documents

Uncertainty in estimation of coalbed methane resources by geological modelling

Description
Description:
Categories
Published
of 14
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Transcript
  Uncertainty in estimation of coalbed methane resources by geologicalmodelling Fengde Zhou  a ,  * , Zhenliang Guan  b a School of Petroleum Engineering, University of New South Wales, NSW 2052, Australia b Key Laboratory of Tectonics and Petroleum Resources of Ministry of Education, China University of Geosciences, 430074, China a r t i c l e i n f o  Article history: Received 25 January 2016Received in revised form31 March 2016Accepted 4 April 2016Available online 19 April 2016 Keywords: Coalbed methane resourcesGeological modellingUncertainty analysis a b s t r a c t This paper presents an uncertainty analysis of coalbed methane resources estimation for a coal seam gas 󿬁 eld containing multiple coal seams. Firstly, logs from nine wells and laboratory data from nine coalseams were used to predict the coal thickness, ash content and gas content at nine boreholes. Secondly,the structural models were determined using the well correlation, structural contour and coal seamthicknesses maps from coal mine data by the convergent gridder and sequential Gaussian simulationmethods. Then, distributions of coal density and ash content were generated in 3D by using sequentialGaussian simulation based on the well log interpretations. Then, the distributions of gas content for eachcell were built by two methods; one is multivariable regression analysis in 3D and the other is sequentialGaussian simulation based on the log interpreted gas content. Finally, coalbed methane resources wereestimated based on the cell volume, coal density, net to gross ratio and gas content. In coalbed methaneresources estimation, four differentdensitycutoffs were used to de 󿬁 ne the net togross ratio of coal in 3D.Results show that the gas contents decreases with increase in depth though with increases of vitrinitere 󿬂 ectance ratio,  󿬁 xed carbon content and pressure. Calculated gas content from linear multivariateregression by using parameters of ash content, volatile matter content and  󿬁 xed carbon content andsample's burial depth matches well with laboratory measured values. The total coalbed methane re-sources estimated are similar by the two geological modelling processes, the multivariable regressionanalysis in 3D and the sequential Gaussian simulation. It has been found that the effects of coal densitycutoffs on coalbed methane resources for different coal seams are different. ©  2016 Elsevier B.V. All rights reserved. 1. Introduction Coalbed methane (CBM) resources Estimation is a complexprocess ( Jenkins, 2008) but it is important for planning and designof coal seam gas production (Zhou et al., 2012). The resource esti-mation is highly uncertain because it depends on the distributionsof coal thickness, coal quality, gas content, gas saturation, etc. Butthe distribution predictions for those parameters are uncertain.Hence, the uncertainty associated with the estimation of CBMresource must be assessed (Zhou et al., 2012, 2015).Three techniques,  󿬁 eld analogs, volumetric methods and prob-abilistic methods were used in CBM resources estimation ( Jenkins,2008). For the volumetric method, the CBM resources arecalculated based on the 3D distributions of coal quality and gascontent. Coal seam thickness, coal density, and gas content areconsidered to be major parameters that introduce the uncertaintyin resource estimation (Wang et al.,1997; Zuo et al., 2009). Hence,the uncertainty in calculatingof CBM resources comes mainly fromtwo sources, well log interpretation and predicted distributions of coalthickness, coal qualityandgascontent(Zhou et al., 2012). Notethat parameters related to 3D modelling, e.g. drilling density,structural uncertainty (well correlation errors, fault geometry,isopach/isochore issues) and gas saturation are also important foruncertainty analysis but beyond the range of this paper.Log interpretation by integrating laboratory and log data iscarried out  󿬁 rstly to estimate the coal quality and gas content atboreholes.Next,thestatisticalanalysisiscarriedouttocharacterisecoal quality and gas content in horizontal and vertical. Thus thegenerated data are used in geological modelling. In laboratory, coalgas content, weight percent of moisture,  󿬁 xed carbon and volatilemattercontent, desorption time and coal bulk densityare normally *  Corresponding author. University of Queensland, School of Earth Sciences, QLD4072, Australia. E-mail address:  f.zhou@uq.edu.au (F. Zhou). Contents lists available at ScienceDirect  Journal of Natural Gas Science and Engineering journal homepage: www.elsevier.com/locate/jngse http://dx.doi.org/10.1016/j.jngse.2016.04.0171875-5100/ ©  2016 Elsevier B.V. All rights reserved.  Journal of Natural Gas Science and Engineering 33 (2016) 988 e 1001  measured foreach sample. Coal maceral compositions, e.g. vitrinitere 󿬂 ectance ratio (VRO, %), vitrinite, liptinite and inertinite, aremeasured in separate experiment. Coal ash and sodium content(Heriawan and Koike, 2008a), volatile matter content (Hagelskamp etal.,1988),sulfurcontent(BancroftandHobbs,1986;Berettaetal.,2010) need to be studied as part of the assessment of coal quality.Thecoalqualityparameterswereused topredictgascontent(Zhouet al., 2012). Nolde and Spears (1998) calculated the gas content using a linear regression equation in which the depth is the onlyindependent variable. According to Nolde and Spears, the gascontent increases with increase in depth. The authors presented amodel equation based on the desorption values for the 61 lowvolatile bituminous coal samples. Fu et al. (2009) estimated the gascontent using multivariable regression analysis by relating data,such as burial depth, resistivity, sonic slowness and density log(RHOB) with measured gas contents from 64 samples. Only one of the  󿬁 ve reported equations by Fu et al. (2009) shows that the gascontent decreases with increase in burial depth but the reasonwasnot represented. Gas content was also estimated by relating withthe pressure, temperate, and the ratio of   󿬁 xed carbon over volatilematter content in wt% (Kim,1977).The log derived data is geostatistical analyzed to estimate dis-tributions of coal thickness ( Jakeman, 1980; Mastalerz andKenneth, 1994), coal quality (Cairncross and Cadle, 1988;Hagelskamp et al., 1988; Liu et al., 2005; Heriawan and Koike,2008b; Beretta et al., 2010; Hindistan et al., 2010) and coaltonnage (Heriawan and Koike, 2008a). Sequential Gaussian simu-lation (SGS) and ordinary Kriging methods are used to predict thedistribution of coal thickness and coal quality (Beretta et al., 2010).Zhou et al. (2012) presented an uncertainty analysis of CBM re-sources using the stochastic reservoir modelling for a CBM  󿬁 eld insoutheast Qinshui Basin from China. They reported that the het-erogeneities in coal seam thickness and coal quality lead toincreasing uncertainty in estimating CBM resources. The authorsalsoreportedthatthedensityvariogramandcoalseamroofsurfacecontribute less to the uncertainty estimation of CBM resources. Butsuf  󿬁 cient data, sound log interpretation models and appropriategeological methods can improve the reliability of resourceestimation.In this paper, the uncertainty in CBM resources estimation isanalyzed fora producing CBM 󿬁 eld containing multiple coal seams.Well logs, laboratory data and the data obtained from coal miningare used to predict the coal structure, coal thickness, coal qualityand gas content. Noteworthy is that the gas content distribution isestimated by using two different techniques: the multivariableregression analysis in 3D and the SGS. The uncertainty is assessed 󿬁 rstly by analyzing the effects of coal density cutoffs on resourceestimation, secondly comparing the resource estimation based onthe distributed gas content bythe multivariable regressionanalysisin3DandtheSGSanalysisand 󿬁 nally,analyzingtherelationshipsof coal bulk density with ash,  󿬁 xed carbon, and volatile matter con-tents. It is assumed that the statistics of gas content, proximateanalysis and logging data from the nine wells can be taken asrepresentative for the study area. 2. Data  2.1. Field data Fig.1showsthelitho-stratigraphicsuccessionofcoalseamXVIIItocoalseam I fromtoptobottom.Thesecoalseams dip gentlywithdip angles ranging from 5  to 15  except in the vicinity of fault andthe seams dip towards the basin centre in general (Saikia andSarkar, 2013). The mining methods for gently dip seam are thesame as in  󿬂 at seams but mining conditions are a little moredif  󿬁 cult ( Jeremic, 1985). The fault's opening ability affect the gassaturation in coal and the dewatering process because the openfault will be a conduit for gas and water  󿬂 ow.Inthestudyarea,logsareusedtocorrelatecoalseamswiththeircharacteristics, namely stratigraphic sequence, thickness, variationof log attributes etc. Fig. 2 shows two cross sections of the ninewells. Results show that the coal seams can be identi 󿬁 ed by typicalvalues of bulk density (RHOB) and gamma ray (GR). The sandstonesequencesarecharacterizedbylowerGRresponse,however,higherRHOB, which is similar to that of the mudstone; GR response inmudstone is the highest of the three. Archaean gneiss is located atthe south of the study area (Paul and Chatterjee, 2011; Singh et al.,2013;SaikiaandSarkar,2013)whichwasintersectedbythewell#F(Fig. 2) at its deepest section. The Archaean gneiss is granite gneissand characterized by high RHOB and low GR.Structural map on the roofs of each coal seam and the thicknesscontour maps of each coal seam were drawn by using coal miningandcoalseamgasboreholes.Fig.3showsthemapsofstructureandcoal seam thickness for one of the coal seam No. III e IV. It showsthat there are 17 faults and 14 of them are normal faults. Threereverse faults are located in tectonic transition area. Coal seam No.III e IV is thicker in the middle and north part of the study area.    L  a   t  e   P  e  r  m   i  a  n   M   i   d   d   l  e   P  e  r  m   i  a  n   E  a  r   l  y   P  e  r  m   i  a  n  Archaean AgeFormation/StagesCoal SeamsLithology    R  a  n   i  g  a  n   j   B  a  r  e  n   M  e  a  s  u  r  e   B  a  r  a   k  a  r Talchir  LohpitiTelmuchiaJamdihMurlidihMahudaHariharpur PetiaShibabudihPhularitandBrariNardkarkiGoladadih    U  p  p  e  r   L  o  w  e  r   L  o  w  e  r   U  p  p  e  r   L  o  w  e  r   U  p  p  e  r   M   i   d   d   l  e Early Cretaceous III IIIIVV/VI/VII VIII IX X XIXIIXIIIXIVXV XVIXVIIXVIII Dolerite dykes Mica lamprophyre dykes and sillsFine grained feldspathic sandstones, shales with coal seams, maximum thickness=725mMassive buff coloured sandstones, grey shale, carbonaceous shale with/without very thin lenses of coal, maximum thickness=625mBuff coloured medium and coarse grained sandstone, grits, shale, carbonaceous shale, siltstone and thick workable coal seams, maximum thickness=1250mGreenish shale, very fine grained sandstone, sandy shale, conglomerate and basal tilliteMetamorphics Fig. 1.  A schematic of the stratigraphic succession showing coal seams. Coal seamsnumbers in rectangles are objectives of this study (After Saikia and Sarkar, 2013). F. Zhou, Z. Guan / Journal of Natural Gas Science and Engineering 33 (2016) 988 e 1001  989   2.2. Laboratory data Asummaryofdesorptionandcoalqualitydataobtainedfrom68coal samples from  󿬁 ve wells (wells No. A, B, C, D and E in Fig. 3) ispresented in Table 1. The desorption data include the total gas content,naturaldesorptiongascontent,losinggascontent,residualgas content and desorption time. Coal quality tests include themoisturecontent,ashcontent, 󿬁 xedcarboncontent,volatilemattercontent and coal density. Table 2 lists the correlation coef  󿬁 cientmatrix for the nine properties of the 68 samples which are listed inTable 1. Fig. 4 shows the variation of raw gas content, moisture content,ashcontent, volatilemattercontent,and 󿬁 xedcarbonwithdepth. Results show that the raw gas content ( Q  ) and volatilematter content (VM) have negative relationships with depth; ashcontent (ASHR) showsa negativeproportionality with 󿬁 xed carbon(FC). Fig. 2.  Twowell correlation sections, (a) Section #S 1  and (b) Section #S 2 . MD in the  󿬁 rst trackis measureddepth in m, RHOB in the second trackis coal bulk density in g/cm 3 , and GR in the third track is gamma ray in API. Distances between wells are not to scale. F. Zhou, Z. Guan / Journal of Natural Gas Science and Engineering 33 (2016) 988 e 1001 990  3. Results  3.1. Log interpretation 3.1.1. Coal quality The proximate laboratory analysis provides the contents of ash,moisture,  󿬁 xed carbon and volatile matter as shown in Table 1.Among these parameters, ash content (wt%) can be related to themeasured coal density. Fig. 5a shows the scatter plot of themeasured coal bulk density and the ash content. The correlationcoef  󿬁 cient (R  2 ) has been found to be 0.74. The equation is:  ASHR ¼ 97 : 67 $ r c   114 : 13 (1) where,  ASHR  is the calculated ash content in wt% as received basis, r c   is the measured coal bulk density in g/cm 3 .In this studythe lab measured density is close tothe log derivedRHOB. Hence, we replace  r c   with  RHOB  in Eq. (1).Fig. 5b shows the scatter-plot of   󿬁 xed carbon content and ashcontent and relationship derived from the regression analysis. Themodel can be expressed as: FC   ¼ 0 : 902   ASHR þ 79 : 056 (2) where,  FC   is the calculated  󿬁 xed carbon content in wt% as receivedbasis.Fig.5cpresentstherelationshipbetweenvolatilemattercontentwith ash content. Results show that therelationship between themis weak. This is because the volatile matter content is not onlyrelated to ash content also to the depth and  VRO  (Amijaya and Fig. 3.  Elevation on the roof of coal seam No. III e IV (a), and thickness contour map of coal seam No. III e IV (b). F. Zhou, Z. Guan / Journal of Natural Gas Science and Engineering 33 (2016) 988 e 1001  991

Molds

Oct 8, 2019

9 quiz 2.3

Oct 8, 2019
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x