Reducing offshore drilling and construction risks with Metocean - - PowerPoint PPT Presentation

reducing offshore drilling and construction risks with
SMART_READER_LITE
LIVE PREVIEW

Reducing offshore drilling and construction risks with Metocean - - PowerPoint PPT Presentation

Reducing offshore drilling and construction risks with Metocean data Mark Calverley R&D manager Fugro Metocean Business Line Society of Petroleum Engineers(SPE) London Evening Programme Meeting 26th April 2016 1 SPE Evening


slide-1
SLIDE 1

www.fugro.com 1

Society of Petroleum Engineers(SPE) London Evening Programme Meeting 26th April 2016

SPE Evening Programme Meeting 26 April 2016

Reducing

  • ffshore drilling

and construction risks with Metocean data

Mark Calverley R&D manager – Fugro Metocean Business Line

slide-2
SLIDE 2

www.fugro.com 2

Outline

SPE Evening Programme Meeting 26 April 2016

  • Metocean – what, why?
  • Drilling
  • Metocean Considerations
  • When to measure?
  • Construction
  • Weather risk
  • Understanding forecasts
slide-3
SLIDE 3

www.fugro.com 3

Winds

SPE Evening Programme Meeting 26 April 2016

slide-4
SLIDE 4

www.fugro.com 4

Waves

slide-5
SLIDE 5

www.fugro.com 5

Currents / Water Levels

Courtesy of NASA's Goddard Space Flight Center

SPE Evening Programme Meeting 26 April 2016

slide-6
SLIDE 6

www.fugro.com 6

Physical Oceanography

  • Temperature
  • Flow assurance
  • Conductivity
  • Density
  • Member weights
  • Sound Profile
  • Survey
slide-7
SLIDE 7

www.fugro.com 7

Engineering / Metocean Interactions

SPE Evening Programme Meeting 26 April 2016

Engineering Schedule Operations Marine Warranty / Legislation

Engineering

Regional Oceanography Metocean Data Requirements Metocean Data Analysis

Metocean

Instrument Choice Sampling Strategy Installation

Measurements

Model Choice Spatio-temporal Grid size Installation

Modelling

slide-8
SLIDE 8

www.fugro.com 8

Structural / Budget Challenges

Drilling Metocean Exploration Engineering design Metocean Strategy Contractors

slide-9
SLIDE 9

www.fugro.com 9

Metocean Strategy across the life cycle

Risk Cost

slide-10
SLIDE 10

www.fugro.com 10

Metocean for Drilling – Operational Planning / Engineering

  • Selection of drilling platform
  • Jackup
  • Generally wave dominated forces
  • Water levels also important
  • Current / wind loading of less

importance

  • Jacking operations most

vulnerable

  • Reliable wave criteria
  • Reliable water level criteria
  • Joint probability of wave/water

levels

  • Knowledge of temporal variability

to optimise jacking operations

slide-11
SLIDE 11

www.fugro.com 11

Metocean for Drilling – Operational Planning / Engineering

Selection of drilling platform

  • Drillship
  • Semi-sub (DP)
  • Semi-sub (anchored)
  • Wave directionality of importance

to drillships

  • Directional differences between

loading important

  • Wave period critical
  • Currents important to station

keeping and riser analysis

  • Wind loading important in areas of

squalls

slide-12
SLIDE 12

www.fugro.com 12

Areas of Uncertainty

  • Hindcast Model reliability (atmospheric, waves,

currents)

  • Analysis Methods
  • Response modelling
  • Metocean processes not represented in models

(temporal or spatially) DNV Guidance note: For meteorological and oceanographic data a minimum of three to four years of data collection is recommended. Planning generally based on hindcast data:

  • Temporal and spatial resolution
  • Validation / verification

Metocean for Drilling – Operational Planning / Engineering

SPE Evening Programme Meeting 26 April 2016

slide-13
SLIDE 13

www.fugro.com 13

Atmospheric Model reliability

Atmospheric Models Assimilation data – helps to define current state of the atmospheric – informed by satellites,

  • bservations (>8,000 Metar/Synop)
slide-14
SLIDE 14

www.fugro.com 14

Wave Model reliability

Wave Models Assimilation data – helps to define current state of the wave field – informed by satellites (altimeter / SAR / GNSS*), observations (order of magnitude fewer than meteorological stations)

slide-15
SLIDE 15

www.fugro.com 15

Current Model Reliability

Current Models Assimilation data – helps to define current state of the ocean circulation– informed by satellites (altimeter / SST), observations (3000 Argo floats)

slide-16
SLIDE 16

www.fugro.com 16

Ensemble Models

Limitations of models recognised

  • Certainly in forecasting, lead to the implementation of ensemble forecasting
  • Needs end user to understand

Probability map for temperature Standard deviation of a temperature ensemble forecast

slide-17
SLIDE 17

www.fugro.com 17

Temporal Considerations

SPE Evening Programme Meeting 26 April 2016

What is the temporal resolution of the model data?

  • 1-hourly
  • 3-hourly
  • 6-hourly

What is the lead time for an operation? What is the duration of an operation? What are the time scales of the Metocean processes? Some processes not well represented in models include:

  • Squalls
  • Solitons
  • Polar lows

Use of complementary data, e.g. satellite data.

slide-18
SLIDE 18

www.fugro.com 18

Model resolution

Spatial Considerations

SPE Evening Programme Meeting 26 April 2016

slide-19
SLIDE 19

www.fugro.com 19

Model resolution

Spatial Considerations

SPE Evening Programme Meeting 26 April 2016

slide-20
SLIDE 20

www.fugro.com 20

Do the models domains extend, at appropriate resolution, to topography / coastlines that might drive the wind forcing?

Spatial Considerations

Can the models capture the spatial scales of the Metocean

  • processes. For example polar lows.

During planning can you rely on the geographic registration of frontal features?

SPE Evening Programme Meeting 26 April 2016

slide-21
SLIDE 21

www.fugro.com 21

Is the environment severe? Can the requirements be addressed by a desk study?

  • Are metocean processes that impact drilling

considerations properly represented?

  • Early commission of a desk study might help to

understand risks. Prospectivity – could site specific measurements add value to fast tracking development? Are data needed in region for other assets? Cost constraints – typically measurements prior to discovery are increasingly rare outside frontier regions such as Barents Sea. Typical year long metocean measurement campaign equivalent to 4 or 5 days downtime!

Model versus Measured Data

slide-22
SLIDE 22

www.fugro.com 22

  • Forecast reliability informed by site specific

measurements

  • Operational decision making
  • Legislative requirements (e.g. CAA CAP437, NTL)
  • Informing future development through site specific

data collection

Metocean for Drilling – Real time support

slide-23
SLIDE 23

www.fugro.com 23

Forecast reliability

Does not provide information of timing of frontal systems.

slide-24
SLIDE 24

www.fugro.com 24

Operational decision making

slide-25
SLIDE 25

www.fugro.com 25

Soliton Early Warning System

Soliton generation zone Andaman Sea

Malacca Strait

Sumatra 2 x Real-time SEWS moorings SEWS#1 SEWS#2

slide-26
SLIDE 26

www.fugro.com 26

Soliton Early Warning System

Warning Level Current Speed (knots) Actions

LOW < 1.5 Record the solitons in daily 24-hour summary, but no warning required or action to be taken by the rig MEDIUM 1.5 to 2.0 Issue soliton warning by email, but the rig will probably not take action HIGH 2.0 to 3.0 Issue soliton warning by email and follow up by calling OIM. The rig will tighten anchor wires and standby VERY HIGH > 3.0 Issue soliton warning by email and follow by calling OIM. Rig will prepare for possible disconnect

slide-27
SLIDE 27

www.fugro.com 27

CAP437

slide-28
SLIDE 28

www.fugro.com 28

Data to support Development

Drilling platforms offer a relatively cheap measurement opportunity. Leverage to collect data to understand reliability of hindcast models supporting engineering, particularly valuable for current models. Provide measurement of processes beyond model capabilities, e.g. squalls. DNV requirement to engineer against measured winds.

slide-29
SLIDE 29

www.fugro.com 29

Hindcast Model reliability (atmospheric, waves, currents) Analysis Methods Response modelling Metocean processes not represented in models (temporal or spatially) Forecast Model reliability (atmospheric, waves, currents) Metocean Awareness Measurement QA

Where do the uncertainties /risks lie?

SPE Evening Programme Meeting 26 April 2016

slide-30
SLIDE 30

www.fugro.com 30

Planning Data to support operational planning. Ways to characterise Metocean conditions for operational planning. Forecast considerations. Transport / Execution Ways to forecast / measure Metocean conditions for

  • perational planning.

Analysis Quantification of downtime, Accuracy of forecasts, etc.

Metocean within construction

SPE Evening Programme Meeting 26 April 2016

slide-31
SLIDE 31

www.fugro.com 31

Weather Cost Exposure

Assume an operation west of Shetland with £250k day rate. 1. Joint frequency distribution based on 4m threshold on significant wave height. 2. Joint frequency distribution based on 4m threshold on significant wave height and peak period below 10s 3. Persistence based on significant wave height and 18 hour duration. 4. Weather windows analysis based on:

  • 18-hour duration
  • Operational threshold = 4m, Tp > 10s; 2.5m, Tp < 2.5s

SPE Evening Programme Meeting 26 April 2016

slide-32
SLIDE 32

www.fugro.com 32

Weather Cost Exposure

SPE Evening Programme Meeting 26 April 2016

slide-33
SLIDE 33

www.fugro.com 33

Weather Cost Exposure

August September Analysis Hs =4m Hs=4m, Tp<10s Hs=4, Durn>18hr Downtime days August 0.45 8.2 0.10 September 3.47 18.62 4.13 Downtime cost August 112,375 2,056,452 24,800 September 868,500 4,656,458 1,031,250 Sep-Aug 756,125 2,600,006 1,006,450 WOW (hours) Weather cost (£) August September Average 1.9 8.0 20,060 83,177 Maximum 54.0 96.0 562,500 1,000,000 P10 6.0 30.0 62,500 312,500 P20 0.0 12.0

  • 125,000

P30 0.0 0.0

  • SPE Evening Programme Meeting 26 April 2016
slide-34
SLIDE 34

www.fugro.com 34

Fatigue potential is high. Directly measure fatigue. Compare Metocean design conditions of transport to those experienced. Identify potential issues for inspection prior to installation.

Transport

SPE Evening Programme Meeting 26 April 2016

slide-35
SLIDE 35

www.fugro.com 35

Forecast Reliability Measurements Response Modelling

Execution

Metocean Awareness

SPE Evening Programme Meeting 26 April 2016

slide-36
SLIDE 36

www.fugro.com 36

Forecast (Model) Reliability

Weather Forecast Level Meteorologist rqd.

  • n site

Independent WF source Maximum WF interval A Yes(1) 2(2) 12 hours(3) B No(4) 2(5) 12 hours C No 1 12 hours

1. There should be a dedicated meteorologist, but it may be acceptable that he/she is not physically present at site. The meteorologist opinion regarding his preferable location should be duly

  • considered. It is anyhow mandatory that the dedicated meteorologist has continuous access to

weather information from the site and that he/she is familiar with any local phenomena that may influence the weather conditions. 2. It is assumed that the dedicated meteorologist (and other involved key personnel) will consider weather information/forecasts from several (all available) sources. 3. Based on sensitivity with regards to weather conditions smaller intervals may be required. 4. Meteorologist shall be conferred if the weather situation is unstable and/or close to the defined limit. 5. The most severe weather forecast to be used. DNV-OS-H101- Marine Operations, General

How is the forecast generated?

SPE Evening Programme Meeting 26 April 2016

slide-37
SLIDE 37

www.fugro.com 37

Are all forecasts equal?

SPE Evening Programme Meeting 26 April 2016

  • How is the forecast generated?
  • Which models support the forecast?
  • What other ‘guidance data’ supports the forecast?
  • Does the forecaster have regional experience?
  • Are observations available to the forecaster?
  • Are the synoptic difficult / easy to forecast?
slide-38
SLIDE 38

www.fugro.com 38

Alpha Factors

α Design Hs = 2m Design Hs = 4m TPOP A B C A B C ≤12 0.76 0.80 0.95 0.83 0.87 1.00 ≤24 0.73 0.77 0.84 0.80 0.84 0.87 ≤36 0.71 0.75 0.77 0.77 0.80 0.80

WF WF WF WF Operation starts TR Required weather window with OPWF = α x OPLIM Estimated time for the operation TPOP (Basis for selecting α-factor) Contingency TC DNV-OS-H101- Marine Operations, General

SPE Evening Programme Meeting 26 April 2016

slide-39
SLIDE 39

www.fugro.com 39

Alpha Factors Reliability

SPE Evening Programme Meeting 26 April 2016

  • No requirement for 2nd forecast to be forecaster driven, many companies use model driven forecasts

to reduce cost.

  • Potential for underlying models to be similar, e.g. public domain models such as WW3 or ECMWF.
  • No consideration of ensemble forecasting
  • Reliability of site specific forecasts not addressed.
  • No consideration of reliability of models and forecasting under different synoptic conditions.
  • No consideration of personnel competency
slide-40
SLIDE 40

www.fugro.com 40

Forecast Reliability – Dichotomous Approach

Result Forecast Hit event forecast to occur, and did occur Miss event forecast not to occur, but did occur False Alarm event forecast to occur, but did not occur Correct negative event forecast not to occur, and did not occur Observed yes No Total Analysis yes hits false alarms forecast yes no misses correct negatives forecast no Total

  • bserved yes
  • bserved no

total

Requires investment in measurement and forecasting prior to construction. Potential to use exploration phase to build this.

SPE Evening Programme Meeting 26 April 2016

slide-41
SLIDE 41

www.fugro.com 41

Measurements - Site specific

Measurements provide:

  • Confidence in forecast model (does T+0 match reality)
  • Potential to reduce DNV α factor
  • Confidence in data during operation and for contractual purposes (weather claims) post execution.
  • Potential to offer early warning systems

SPE Evening Programme Meeting 26 April 2016

slide-42
SLIDE 42

www.fugro.com 42

Measurements - Spatial Technologies

Measurements can provide a spatial picture.

  • Vessel Mounted Acoustic Doppler Current Profilers (4 knot limit)
  • Airborne current measurement systems (ROCIS uses optical imagery to provide surface current

data)

SPE Evening Programme Meeting 26 April 2016

slide-43
SLIDE 43

www.fugro.com 43

Response Modelling

Joe Bloggs

SPE Evening Programme Meeting 26 April 2016

slide-44
SLIDE 44

www.fugro.com 44

Structural Monitoring

Measuring Environmental Forces provides one half of the equation. The other half is the motion response and structural response Consideration should also be given to measuring structural responses during construction. Examples include:

  • Motion sensors
  • Not a single sensor but at critical locations,

e.g. crane tip, launch points, etc

  • Strain measurements

Likely driven by critical components/operations.

SPE Evening Programme Meeting 26 April 2016

slide-45
SLIDE 45

www.fugro.com 45

Metocean Awareness

The data consumer! Convert data into decisions!

SPE Evening Programme Meeting 26 April 2016

slide-46
SLIDE 46

www.fugro.com 46

Metocean Awareness

Parameter (s) Operating limit Status Hs Tp Ws Cs WL

The data consumer! Convert data into decisions!

SPE Evening Programme Meeting 26 April 2016

slide-47
SLIDE 47

www.fugro.com 47

Execution Summary

Understand forecast provenance / reliability. Complement forecasts with measurements to reduce risk. Consider use of spatial measurement techniques to identify frontal features. Consider use of remote measurements to provide early warning.. Help to drive new technology adoption –

  • perators/contractors often very conservative.

Move to decision making systems rather than Metocean data systems. Incentivise risk management!

SPE Evening Programme Meeting 26 April 2016

slide-48
SLIDE 48

www.fugro.com 48

Analysis

Plan Execution Forecast performance Operational Performance Lessons Learned Contractors

SPE Evening Programme Meeting 26 April 2016

slide-49
SLIDE 49

www.fugro.com 49

Technology Helping Hands

Marine Autonomous Systems offer a low cost way to collect spatial measurements

  • Market movement
  • Regulatory framework being established (MAS RWG)
  • Cost driven market (analagous to diver to ROV)
  • System includes analysis and delivery of relevant information to end user.
  • Low power sensor / increasing power capability

SPE Evening Programme Meeting 26 April 2016

slide-50
SLIDE 50

www.fugro.com 50

Technology Helping Hands

SPE Evening Programme Meeting 26 April 2016

Designing for data generation and capture Data generation and capture Data storage, access Big data modelling and analytics Visualising big data, models and analysis Data driven insights Data driven decision making under uncertainty Monitoring and evaluation

Big Data Wheel

slide-51
SLIDE 51

www.fugro.com 51

Summary

  • Create an open dialogue with Metocean experts.
  • Evaluate Metocean based risks with respect to operations / engineering design through workshops

rather than just reports.

  • Consider cost benefit of different Metocean strategies (model versus measured, etc) with full

cognisance of risks.

  • Ensure timely consideration is provided.
  • Provide Metocean awareness training to staff (for example IMarEST Metocean Awareness Course)
  • Include Metocean in project stage gates.

Remember Chaos Theory was introduced by a Meteorologist (Edward Lorenz) in his 1972 paper entitled "Predictability: Does the Flap of a Butterfly's Wings in Brazil Set Off a Tornado in Texas?"

slide-52
SLIDE 52

Thank you. Questions?

Mark Calverley M.calverley@fugro.com 01491 820546