Allocation of Unidentified Gas Statement

Allocation of Unidentified Gas Statement16th May 2011

Outline

• Introductions• Presentation of the AUGS

• High level plan• Consultations/presenting to industry• Allocation of Unidentified Gas

• Understanding of the problem• Proposed methodology

• Summary / Q&A

• Future Meetings• AOB

What is GL?GL is a world class technical assurance and consulting company and leading ship classification society.

Three business areas:•Maritime Services•Renewables – GL Garrad Hassan•Oil And Gas – GL Noble Denton

GL Noble Denton - Our HeritageOriginally formed as the Research and Development Group of British Gas

With the re-structuring of British Gas our name changed to BG Technology and acquired Transco Engineering Services & Transco Technical Training in 1997

In 2001 we acquired Risx and Stoner Associates Inc. and became Advantica

In 2007 Advantica was acquired by Germanischer Lloyd

GL Noble Denton has become a leading global advisor to the Oil & Gas industryOil & Gas assets

MOPU

Offshore Structures

Onshore Facilities

MODU

Pipelines

Subsea Installations

3,000 Employees in35 Countries

CanadaUSAMexico

UK

Poland

TunisiaLibyaEgyptPeru

Brazil Saudi Arabia

ChinaKorea

BruneiIndonesiaAustraliaMalaysia

ThailandPhilippinesVietnamSingapore

Russia

Italy

UAE

Kasachstan

IranPakistan

India

Qatar

GL Noble Denton comprises experts from:Technical Assurance Verification, Certification and

Classification Industrial Inspection

Project Execution Project Management Design Transportation & Installation Due Diligence Construction Monitoring

Competencies along the whole asset life cycle

Marine Consulting & Operations Marine Warranty Marine Consulting, Vetting &

Surveys, Marine Operations Marine Casualty Investigation Dynamic PositioningAdvanced Engineering & Consulting

Adv. Analysis & Eng. Consulting Field Development Planning Testing Services Asset Integrity Mgmt / Optimization Safety & Risk Software Solutions

Trinidad

Norway

Germany

Kuwait

Bahrain

Oman

Conflicts of Interest• GL Noble Denton is independent of all UK industry bodies

• GL Noble Denton have not previously been involved with this particular unidentified gas problem

• One of our core values is to be “the most respected international technical advisor and trusted partner” for all our clients

• Our objectives are simple – to devise a methodology that is unbiased, meets the requirements of the industry bodies and is approved by the UNCC

Overview of Project Plan

High Level Plan

Presenting to the UNCCThe AUGS covers• Adherence to AUGE Guidelines terms of reference• Proposed topics for unidentified gas and their status• Data provision status/issues• AUGE Methodology

• Reasoning • Assumptions • Data Used• Issues/recommendations going forward• For incomplete aspects of the methodology (e.g. pending data) the AUGS contains details of

the remaining steps to complete it

• Consultation questions and responses as appropriate

Reporting Progress

• Primarily through the main UNCC presentations and publications of AUGS

• There is also visibility of activities from data requests and responses to consultation questions

• In addition, brief progress reports will be provided via JO website

Consultations

Industry Bodiesissues/questions

GL Noble Denton

Industry Bodies issues/questions

• Collate• Compare vs criteria• Investigate / assess• Generate responses

GL Noble Dentonpresent answers via • AUGS • Industry meetings• JO website

• Update methodology• Update AUGS

Questions/issues can be raised via [email protected]

Unidentified Gas – The Problem

Unidentified Gas – Understanding of the Problem

The AUGE• Robust and repeatable algorithm• Estimates of Unidentified Gas that are

• Accurate• Unbiased

• Unidentified Gas estimation is the core of the consultation/publication process

Unidentified Gas – Statement of the Problem

UK gas market sectors• Daily Metered (DM)• Non-Daily Metered Larger Supply Points (NDM LSP)• Smaller Supply Points (SSP)

• Threshold between SSP and LSP is 73,200kWh pa


Additional factors:

• LDZ Shrinkage• Leakage• Own Use Gas• Transporter-Responsible Theft

• Unidentified Gas

• Both unknown, both have to be estimated


Unidentified Gas• Unregistered and Shipperless sites• Independent Gas Transporter CSEP Setup and Registration Delays• Corrections to the shrinkage estimate• Shipper-responsible theft

• Potential additional sources of Unidentified Gas• LDZ offtake meter errors• Supply point meter errors• These only apply if there is a long-term metering bias


Calculation of Unidentified Gas• Understand process flow

• Location of UG at each stage of the calculation

• Calculate UG at optimal process stage• Directly• By subtraction

• Time frame covered by individual UG calculations• Daily (aggregate results to annual level for UNC table)• Monthly (aggregate results to annual level for UNC table)• Annual


Previous UNC Modification Proposals• 194/194A

• Gave framework for quantifying Unidentified Gas• Relied on set percentages of RbD or set values

• 228/228A• Populated framework with set percentages (228) or set values (228A)• No provision for year-on-year updates at this stage

• Potential areas for improvement• Both carry assumption that all or most of RbD is Unidentified Gas• Recognition of allocation algorithm error


• The AUGE is independent and impartial

• Modification 229 allows UG estimates to be updated every year• No UNC amendments required

• Estimates from the AUGE circulated and agreed prior to implementation

Unidentified Gas – The Solution


Daily Allocation Process• Based on End User Categories• Trained on metered data

• Minimum 3900 Small NDM, 1600 Large NDM supply points• 2010 analysis actually used a total of 12715 supply points

• Estimates raw (metered demand) – i.e. no Unidentified Gas• Scaled up to meet LDZ load (minus Shrinkage and DM load)• Scaling process introduces Unidentified Gas

• Value unknown• Split between EUCs by AQ


Composition of allocation estimate• Actual EUC load• Model error (can be positive or negative)• Unidentified Gas (always positive)

• Each EUC total known, but composition unknown


Final calculation steps• Aggregate to SSP and NDM LSP level

• SSP corresponds to one EUC• NDM LSP is the sum of the remaining EUCs• This is the basis for initial (deemed) load allocation

• Actual NDM LSP (monthly) meter reads become available• Apply Reconciliation by Difference

• NDM LSP load now accurate• Model error eliminated• SSP allocation is now actual SSP load plus Unidentified Gas• SSP/UG split unknown

Unidentified Gas – The SolutionAllocation Composition - Individual EUCs

0

50

100

150

200

250

300

350

1 2 3 4 5 6 7 8 9

EUC

TWh

EUC Load Model Error + Unidentified Gas

Allocation Composition - By Sector

0

50

100

150

200

250

300

350

SSP LSP NDM

TWh

EUC Load Model Error + Unidentified Gas

Allocation Composition - Post RbD

0

50

100

150

200

250

300

350

SSP LSP NDM

TWh

EUC Load Unidentified Gas


Unidentified Gas Estimation• Location known, UG accumulated into one figure• Estimate should now be made• Recommended approach – calculate each component directly• Calculation granularity is part of analysis

Unidentified Gas – Individual Elements

Unidentified Gas – Data Requirements• Site AQs (for all DM, LSP and SSP sites)• Actual daily loads (DM sites)• Profiled daily loads (from meter reads for LSP NDM sites)• Deemed daily loads (LSP NDM and SSP sites)• RbD quantities• LDZ Shrinkage (calculated)

Plus data for• Unregistered and shipperless sites • IGT CSEP setup and registration delays• LDZ shrinkage initial and final estimates• Theft (including information on anti-theft initiatives)

Unidentified Gas – Unregistered/Shipperless Sites

• Unregistered Sites• Opening meter reading taken when site is registered• If this is not zero, Unregistered gas has been taken

• Shipperless Sites• Closing meter reading taken when site is unregistered• Opening meter reading taken when site is registered• If they differ, Shipperless gas has been taken• This gets assigned to SSP sector via RbD

• Potential impact of Mod 369

Unidentified Gas – Unregistered/Shipperless Sites• Unregistered Sites – Shipper Activity/Orphaned

• Data has been requested splitting these into “meter”/”no meter” categories• Determine proportion with meter that actually flow gas before they are registered• Analyse all opening meter readings to derive percentage non-zero• Apply this proportion to “meter” group• All sites in this group are more than 12 months old, hence unregistered consumption = AQ

• Unregistered Sites – Sites Created <12 Months• Proportion flowing gas the same as for Shipper Activity/Orphaned sites• Assume that rates of new sites getting MPRNs and sites with MPRNs becoming registered

are both steady• Sites will, on average, have been taking gas for half the time period in question• Report currently semi-monthly, hence average time taking gas is one month

Unidentified Gas – Unregistered/Shipperless Sites• Shipperless Sites – Passed To Shipper

• Only sites visited after being shipperless for more than 12 months are reported• More sites will have been shipperless for less than 12 months

- those flowing gas are still contributing to Unidentified Gas• Relevant data requested from Xoserve

- all shipperless sites (including <12 months) - all >12 month sites visited (not just those found to be flowing gas)

• Calculate “still flowing gas” rate from site visit data• Apply to <12 month sites• Carry out final calculation

- all >12 month sites in report have been flowing gas for full 12 months - the <12 months sites still flowing gas will have been doing so for an average of 6 months

• Shipperless Sites – Shipper Specific Report• Calculations as above

Unidentified Gas – Unregistered/Shipperless Sites• No Activity

• Sites currently being processed• These will end up in one of the other categories• Split between those categories pro-rata before calculations carried out

• Legitimately Unregistered• These sites are believed to have no meter• Hence they do no contribute to Unidentified Gas

Unidentified Gas – IGT CSEP Registration Delays

• Within CSEPs, sites cannot take gas without being registered• UG from this source comes from unregistered CSEPs only

• Entire network not registered on Xoserve system

• Regular meetings held with IGTs• Report on statistics

• 32000 CSEPs• Approximately 1200 are currently unregistered• Down from 3000 in 2009

• Get average CSEP composition from known (registered) CSEPs• Apply to unknown population to estimate consumption

Unidentified Gas – Shipper-Responsible Theft

• Data available for• Detected theft• Alleged theft

• These are high and low limits for true theft• Two-yearly meter inspections ensure that unknown theft (ie theft that exists but has

never been alleged) is limited

• Changes in theft detection rates when shipper initiatives take place can be used to estimate likely value of true theft between upper and lower limits

• It is recognised that true theft is very difficult to pinpoint• Final value will be one agreed to be reasonable by all parties

Unidentified Gas – Shrinkage Error

• Difference between initial estimate and final estimate of shrinkage• Can be positive or negative

• Shrinkage calculated as defined in UNC• Most accurate models available are used

• National Leakage Tests• Own Use Gas model• Standard Theft assumptions

• Final shrinkage values recalculated after the event and RbD credited• Hence error of initial (pre-year) estimate already accounted for in process

Unidentified Gas – Meter Error

Metering takes place• At LDZ offtakes• At Supply Meter Points

• DMs – daily• NDM LSP – usually monthly• SSP – less frequently

• Meter errors only have an impact over time if they carry a long term aggregate bias• Potential effects can be positive or negative• If present, this is a correction to RbD rather than Unidentified Gas

Unidentified Gas – Meter Error

• SSP meters use a rubber diaphragm• These can warp over time• They are set up to be unbiased on average over their 20-year lifespan• SSP metering does not affect Unidentified Gas calculation anyway

• LSP and larger meters use metal components• Regular maintenance• No warping over time• Therefore no long-term bias from this source

• Meter error can be discounted from analysis• Backs up conclusions drawn in previous UNC Mods

Unidentified Gas – Unknown Sites

• Sites that are taking gas that• Are not registered with a shipper• Do not even have a MPRN

• These are unknown to the Xoserve system

• Data has been requested on this subject

Summary – Q/A

• Any other areas of UG to consider?

• Clarifications/Questions?

• Can raise further questions/issues post meeting via [email protected]

Thank you for your attention

Documents

Allocation of Unidentified Gas Statement