Docstoc

Interactive Information System For Online Processing Geo-Technological Data (GTD) Sinking Wells

Document Sample
Interactive Information System For Online Processing Geo-Technological Data (GTD) Sinking Wells Powered By Docstoc
					                                                             (IJCSIS) International Journal of Computer Science and Information Security,
                                                                                                                 Vol. 9, No. 5, May 2011



          Interactive Information System for online processing geo-
                    technological data (GTD) sinking wells
                                                    Information Systems
                                                          Safarini Osama
                                                          IT Department
                                                        University of Tabuk,
                                                            Tabuk, KSA
                                                     usama.safarini@gmail.com
                                                        osafarini@ut.edu.sa


Abstract—: Online management of drilling requires the choice of                -       Obtaining of preliminary “integral” well logging
an informed decision of many possible because of the volume of                 curve and its segmentation.
incoming and processed GTD, problem arising in the functioning
through management situations. The importance here is the                 From measures of similarity (see Table 1) is selected “distance
information management process to enable effective        man-            indices” similar to a distance by Hamming and Euclid as the
machine decision. So the purpose of work is to Develop a
                                                                          most widespread [2]. The features that describe distance
methodology, algorithm and program for processing
(compression and classification) GTD sinking wells, confirming            indices in this case will be an amplitude and depth, while
the geological GTD, for example, marks mining drill bits;                 measures of similarity – their functions or as analogs of a
                                                                          distance by Hamming or Euclid:
                                                                          -             A product of a module of difference of
Keywords- Man-machine decision, Geo-technological data,                   amplitudes;
classification, compression, correlation, measures of similarity,         -             A product of a module of amplitude difference
marks mining drill bits, data mining geology, Information                 by a difference of depths;
Component, Euclidean and Hamming Distance.                                -             A product of a module of amplitude difference
                                                                          by a square of differences of depths.
                     I.         INTRODUCTION
                                                                          In these segmentation methods a number of segments,
The Work describes methods and means of information                       measures and functions applied here using the program shown
system software for decision-making by the results of geo-                in (Fig .1) can be varied with a possibility to present areas of
technological data (GTD) on bore-hole drilling, compression               segments, their models specifying borders, intersections, etc.
of GTD on a drilling regime, improvement of interactivity of              very close to that which is now assumed for processing of
GTD processing and online management process of drilling,                 vague sets as the measures of similarity of objects, classes are
algorithms of GTD segmentation, a program product, results                the values of the function that belongs to [3].
of GTD processing.


                          II.    DISCUSSION
While classifying GTD, the process of segmentation aimed at
taking On-line decisions in drilling, a forecast of the beginning
or end of an interval, following and prediction of a working
period of a drilling bit, evaluation of wear of drilling tools,
prevention of emergency situations, breaking of equipment
and others [1].
The results of the proposed segmentation provide us a
geological situation through a well depth. The proposed
methods assume interactive interpretation of segmentation and
compression of GTD and a possibility of additional verifying
repetitions and variations. This is connected with division of
GTD into segments, their verification by the identified models
applying two, essentially different methods:
      -       Separately for each well logging curve with their                     Fig.1 Program Interface for classification into classes
      subsequent superposing for final segmentation;



                                                                    106                                  http://sites.google.com/site/ijcsis/
                                                                                                         ISSN 1947-5500
                                                                                                           (IJCSIS) International Journal of Computer Science and Information Security,
                                                                                                                                                               Vol. 9, No. 5, May 2011


                                                                                                                                presented as a much simpler function with the same
    Classification by various measures of similarity          Table 1                                                           characteristics as the original sample.
      Formula of a similarity
             measure          Division of information components
                                 Class №1         Class №2      Class №3
                                                                                                                         •      Classification of each geo-measured data properties.
    1)         RIJ
                        ,
                                                   1,3,4,5,6,7,8      11                 2,9,10,12,13,14
          RI + RJ − RIJ

         1               Rij                     5,3,4,6,7,8,9,11   14,13              2,1,10,12                           1.000
    2)     RIJ +                       
         r
                 r + 1 − ( RIJ + Rij ) 
                                                                                                                         -0.514            1.000
                                                                                                                           0.315

    3)
            RIJ                                    1,3,4,5,6,7,8      11                 2,9,10,12,13,14                  -0.336            -0.135          1.000
       RIJ + RIj + RiJ                                                                                                    -0.498            -0.483

                                                                                                                          -0.573            -0.117           0.651             1.000
    4)
               2 RIJ                               1,3,4,5,6,7,8      11                 2,9,10,12,13,14                  -0.554            -0.593           0.789
         2 RIJ + RIj + RiJ
                                                                                                                             0.435          0.288           -0.716             -0.924               1.000
                                                                                                                             0.504          0.608           -0.746             -0.847
              RIJ + Rij                            5,3,4,6,7,8        1                  14,2,9,10,11,12
    5)                                                                                                                       0.627          0.175           -0.575             -0.805               0.848   1.000
         r + RIj + RiJ
                                                                                         ,13                                 0.524          0.517           -0.473             -0.558               0.543

                                               1
                                                   5,3,4,6            14,9,10,11,13      1,2,7,8,12
                     S
                                           h
    6) dij = ∑Ck (xki − xkj )h 
                 k=1                                                                                                               Table 2.Correlation of parameters with / without separating into
             1                                                                                                                                                   layers
    Cij =
          1 + d ij


    7)   ∑x
          k
                    ik     − x jk                  3,1,4,5            14,2,10,12         13,6,7,8,9,11
                                                                                                                                           Average Test 1 Test 2 Test 3 Test 4 Test 5 Test 6


    8)   ∑k
                xik − x jk                         3,1,2,4,5          14,9,10,11,12      13,6,7,8                                Layer 1    0.000   0.000   0.000   0.000   0.000   0.000   0.000

                                                                                                                                 Layer 2    0.289   -0.027 -0.585 -0.210 0.425 -0.243 -0.246

         ∑ (x
          k
                 ik       − x i )(x jk − x j )     5,4,6,7,8,9,11     14,2,10,12,13      3,1                                     Layer 3    0.169   -0.037 0.171    0.093 -0.131 -0.195 -0.386

    9)                                                                                                                           Layer 4    0.193   -0.030 -0.302 0.621     0.007 -0.045 0.152
                          σ iσ j
                                                                                                                                 Layer 5    0.274   -0.139 0.928 -0.109 0.416       0.000 -0.053
           r−s       s~                            5,3,4,6,7,8        14,2,9,10,11,12,   1
    10)        C ij + Cij                                                                                                        Layer 6    0.233   0.217 -0.911 0.015 -0.253 0.000         0.001
            r        r                                                13                                                         Layer 7    0.283   -0.348 0.911 -0.015 0.253 -0.165 -0.006

                                                                                                                                 Layer 8    0.284   0.286 -0.827 -0.123 -0.374 -0.026 -0.066

                                                                                                                                 Layer 9    0.262   -0.110 0.056    0.518   0.571 -0.313 -0.006

                                                                                                                                Layer 10    0.073   -0.043 -0.003 -0.043 0.017      0.075   0.256

                                     Table 1.Classification by various measures of similarity                                   Layer 11    0.123   0.036 -0.227 -0.362 0.005 -0.087 0.024

                                                                                                                                Layer 12    0.326   -0.806 0.791    0.069   0.064   0.071   0.158

         GTD Processing in two stages

-         In the first stage compression and classification of GTD for                                                                Table 3.Correlation with the marks of mining bits
         each of the measurements. The results are a set of features for
         the second phase.

-        At the second stage, the final classification on the full range of
         GTD, this allows assessing the correlation with marks of bits,                                                                              III.             CONCLUSION
         or data mining geology.
                                                                                                                     The developed information System is an instrument for
                                                        Data Compression                                             decision-making in complicated multi-factor non-formalized
                                                                                                                     cybernetic systems with a feedback, i.e.:
         Stage data compression involves the following steps:
                                                                                                                         •      in interactive assessment of informational significance
                          •          Calculate the autocorrelation function Kxx for every                                       of drilling factors provided by readings of on-land
                                     geo-measured properties curve.                                                             facilities, telemetric and feedback data [4];

                          •          Determine the Tk - correlation interval for each                                    •      support      of     processing  (compression     and
                                     sample. It is determined based on type of                                                  classification) of well sinking results verifying
                                     autocorrelation function.                                                                  geological prospecting data, for instance, on a mark
                                                                                                                                of drilling bit run;
                          •          Approximation of each sample geo-measured
                                     properties sampling interval in depth equal to the Tk.                              •      Developed are algorithms and programs for
                                     In this case geo-measured data properties are                                              segmentation of GTD with a possibility of an
                                                                                                                                interactive assessment of a segmentation quality,




                                                                                                               107                                                  http://sites.google.com/site/ijcsis/
                                                                                                                                                                    ISSN 1947-5500
                                                                         (IJCSIS) International Journal of Computer Science and Information Security,
                                                                                                                             Vol. 9, No. 5, May 2011
          variation     of    a     number       of    segments,
          representativeness, correlation to a geological profile,
          borders of formations, wear of drilling bits, and
          prevention of emergency situations.
                                                                                                                              AUTHOR’S PROFILE
     •    Application of MS Excel for estimation of segments
          of GTD on a drilling regime;                                                                           Dr. Safarini Osama Ahmad Salim had finished
                                                                                                                 his PhD. from The Russian State University of
     •    As seen from Table 2, the correlation shows better                                                     Oil and Gaz Named after J. M. Gudkin,
                                                                                                                 Moscow, 2000,        at    Computerized-Control
          results when separating the well profile into layers;                                                  Systems Department. He was awarded by his
          this reflects the fact of geology changing properties.                                                 participation in Interpretation of measurement
                                                                                                                 data in gas wells, Abstracts of paper presented to
     •    As seen from Table 3, Correlation with the marks of                                                    the third All- Russia Conference of young
                                                                                                                 scientists, specialists and Students on the
          mining bit confirms the changes in geo-measured                                       problems in gas industry in Russia “New technologies in the gas
          data properties or as different layers.                                               industry”, Moscow 1999, 28-30 September.



                                                                                      He obtained his BSC and MSC in Engineering and Computing Science from
                                                                                      Odessa Polytechnic National State University in Ukraine 1996. He worked in
                                REFERENCES                                            different countries and universities. His research is concentrated on
                                                                                      Automation in different branches Specially Oil and Gaz.
[1] Levitzky A.Z., Komandrovsky V. G., Safarini Osama
    Methods and Means to Develop an Information System for On-Line
    Control of Drilling, Scientific-Technical Research Journal, “Automation
    Telemetry and Communication in the Oil Industry”, N 3 2000, PP 7-11.

[2] Safarini Osama
    "Enhanced Decision-Making Computer-Aided Methods for On-Line
    Control of Well Drilling", Abstracts of paper of the IPSI Conference Held
    in Carcassonne, France, UNESCO Heritage, April 27 to 30, 2006.

[3] Levitzky A.Z., Komandrovsky V. G., Safarini Osama
    On Automation for On-Line Control of Well Drilling, Scientific-Technical
    Research Journal, “Automation Telemetry and Communication in the Oil
    Industry”, N 3-4 1999, PP 2-8.

[4] Komandrovsky V. G., Safarini Osama
    On classification of information components of On-Line control of a
    drilling Process, Abstract of paper of the Third Scientific Technical
    Conference, “Urgent Issues of the Condition and Development of the Oil
    and Gas Complex in Russia”, Moscow, 1999 27-29 Jan.




                                                                                108                                   http://sites.google.com/site/ijcsis/
                                                                                                                      ISSN 1947-5500