WITSML Log Enhancement for Wireline Data - PowerPoint PPT Presentation

1 / 22
About This Presentation
Title:

WITSML Log Enhancement for Wireline Data

Description:

History (cont'd) POSC WITSML SIG is formed. POSC takes on role in ... A fast channel' is a channel which is recorded at a sample rate higher than the ... – PowerPoint PPT presentation

Number of Views:144
Avg rating:3.0/5.0
Slides: 23
Provided by: jcthe
Category:

less

Transcript and Presenter's Notes

Title: WITSML Log Enhancement for Wireline Data


1
WITSML Log Enhancement for Wireline Data
  • Jim Theriot, POSC
  • 16 Nov 2004
  • Jim.Theriot_at_POSC.org
  • 1 713 267 5109

2
Log Data FormatsHistory
  • LIS, DLIS
  • comprehensive support for wireline data (/-)
  • binary formats
  • LAS
  • text format
  • simplified log model
  • WellLogML v1.0
  • text format (XML)
  • simplified log model

3
History (contd)
  • POSC began work on WellLogML v2.0
  • Requirement fully support wireline data
  • WITSML developing Log object
  • POSC committed to log format compatibility
  • WellLogML v2.0 put on hold
  • WITSML v1.2 released
  • text format (XML)
  • simplified log model

4
History (contd)
  • POSC WITSML SIG is formed
  • POSC takes on role in maintenance of WITSML
    objects
  • POSC envisions extended scope of WITSML
    objects, architecture across EP
  • POSC proposes enhancements for wireline data

5
New Features
  • Multipass log model
  • Fast channels efficiently supported
  • Array data efficiently represented
  • Dimensionality of array data characterized

6
Terminology What is a log, anyway?
  • pass set of curves, continuously recorded
    created by logging instrument and acquisition
    process
  • log set of passes, packaged as deliverable of a
    logging job created by system at wellsite
  • composite log merged result of multiple logging
    jobs created by technician in office pass and
    log context are minimized

7
Log (loggers view)
Wellbore (context)
Job (event)
log
service type
Pass (event)
curve type
curve
data
8
Log (simplified view)
wellbore
log pass
service type
curve type
curve
data
9
Composite Logsupress artifacts job, pass
wellbore
log
service type
curve type
curve
data
10
Single Pass Log schema
logs
log
logHeader
indexType, startIndex, endIndex, direction, indexC
urve
logData
logCurveInfo
data
11
Multipass Log schema
logs
id, pass, indexType, startIndex, endIndex, directi
on, indexCurve
log
logHeader
block
channel
logData
logCurveInfo
blockId
level
12
Fast/Slow ChannelsWhat is a fast channel?
  • A fast channel is a channel which is recorded
    at a sample rate higher than the nominal sample
    rate of the concurrently recorded samples
    (usually an integer multiple of the nominal
    sample rate)
  • Problems with simple representation
  • Excessive nulls
  • Cumulative indexing errors

13
Array channels(representational)
  • Simple channel
  • atomic value at each depth
  • e. g. X(6600.0 ft) 2.15 gm/cc
  • Array channel
  • array value at each depth
  • e. g. Y(6600.0 ft) 1.3,1.6,1.4,2.1,4.3mD

14
Array channels(logical)
  • Simple channel S fcns(depth)
  • Array channel A fcna(depth,)
  • 1D A1 fcna1(depth,pty_a)
  • 2D A2 fcna2(depth,pty_a, pty_b)
  • 3D A3 fcna3(depth,pty_a, pty_b, pty_c)

15
Array examples
  • formation acoustic slowness fcn(depth)
  • formation acoustic response fcn(depth,
    time)
  • formation acoustic response, directional
    fcn(depth, time, azimuth)

16
Array Dimensionality
  • What is the positional semantics of the array?
  • X2.1, 2.4, 2.7, 2.3, 2.6, 2.5 gt ???
  • Ordered (by rank) list of dimensions
  • Name, description, count, values,
  • nameazimuth count6 values0,60,120,180,240,30
    0(deg)
  • gtX120(deg)2.7

17
Array Definition Schema
logCurveInfo
typeLogData
arrayDef
encodingDef
count, name, propertyType, uom
axisDef
axisValues
18
Curve Data Schema
blockId
level
data
channelIndex
DoubleArray
Base64Array
channelIndex
19
Base64 encoding
  • Similar to hexadecimal
  • Internet standard for encoding of binary data
    IETF RFCs 1421 2045
  • encode 6 bits as A-Z,a-z,0-9,,/
  • efficient representation expansion factor of
    1.33 compared to binary

20
Conclusions
  • Discussed by WITSML group in Paris
  • Recommendation
  • Retain current WITSML Log object
  • Create new object with enhanced features
  • If/when new object is taken up, decide whether to
    keep both objects or drop the current object

21
Conclusions
  • What about WellLogML?
  • deprecate?
  • freeze at v1.0?
  • develop to distinct requirements?
  • wait for WITSML Log enhancements takeup?
  • Inviting suggestions and recommendations
    (especially from WellLogML early adopters!)

22
Questions?
  • Jim Theriot, POSC
  • 16 Nov 2004
  • Jim.Theriot_at_POSC.org
  • 1 713 267 5109
Write a Comment
User Comments (0)
About PowerShow.com