Research ArticleAnalysis and Comparison of Long-Distance Pipeline Failures

Lianshuang Dai12 DongpoWang1 TingWang3 Qingshan Feng2 and Xinqi Yang1

1School of Materials Science and Engineering Tianjin University Tianjin 300072 China2PetroChina Pipeline Company Langfang Hebei 065000 China3PetroChina Pipeline RampD Center Langfang Hebei 065000 China

Correspondence should be addressed to Dongpo Wang wangdptjueducn

Received 20 February 2017 Revised 17 April 2017 Accepted 3 May 2017 Published 25 May 2017

Academic Editor Merv Fingas

Copyright copy 2017 Lianshuang Dai et alThis is an open access article distributed under the Creative CommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

The analysis results of long-distance oil and gas pipeline failures are important for the industry and can be the basis of risk analysisintegrity assessment and management improvement for pipeline operators Through analysis and comparison of the statisticalresults of the United States Europe the UK and PetroChina in pipeline failure frequencies causes consequences similaritiesand differences of pipeline management focusing points and management effectiveness are given Suggestions on long-distancepipeline safety technology and management in China are proposed

1 Introduction

It is important to maintain high-pressure oil and gas pipelinesystems safety and reliability because the products are haz-ardous and may result in fire explosion and poisoning andlead to significant economic losses casualties and environ-mental pollution [1ndash3] By collecting and analyzing the failuredata pipeline operators can find out the causes of failureevents and understand the weak point in pipeline manage-ment which is significant for pipeline risk identificationintegrity assessment riskmitigation and accident prevention[4ndash6]

Statistical results of the US Europe the UK and Petro-China in pipeline failure frequencies causes and conse-quences are comparatively analyzed Similarities and dif-ferences of pipeline management are given Suggestions onlong-distance pipeline safety technology and management inChina are proposed

2 Analysis and Comparison

Failure statistical results of PHMSA in the US [7] EGIGin Europe [8] UKOPA in UK [9] and PNGPC in Chinaon long-distance pipeline failure frequencies causes andconsequences are comparatively analyzed Table 1 shows thepipeline types that the analysis involves

21 PHMSA Failure data of all pipelines in the US is updatedto show the recent 20 yearsrsquo statistical results and detailedinformation by PHMSA Significant incidents of onshorepipelines (for liquid only crude oil and refined andorpetroleum product are involved for gas only transmissionline is involved) are filtered from the database and calcu-lated for the failure frequencies in this paper Significantincidents are those including any of the following condi-tions

(1) Fatality or injury requiring in-patient hospitalization(2) $50000 or more in total costs measured in 1984

dollars(3) Highly volatile liquid releases of 5 barrels or more or

other liquid releases of 50 barrels or more(4) Liquid releases resulting in an unintentional fire or

explosion

211 Failure Frequency Figure 1 shows that from 2004 to2015 failure frequencies of oil pipelines in the US varybetween 04 timeskkmsdotyr and 06 timeskkmsdotyr which isslightly increasing in the last 5 years as seen in the trend lineAs for the natural gas pipelines shown in Figure 2 the numbergoes up from 004 timeskkmsdotyr to 014 timeskkmsdotyr withvibration

HindawiJournal of Petroleum EngineeringVolume 2017 Article ID 3174636 7 pageshttpsdoiorg10115520173174636

2 Journal of Petroleum Engineering

Table 1 Mediums included in the statistical data

NO OrganizationCompany Pipeline MediumCrude Oil Product Oil Natural Gas

1 PHMSA

2 EGIG

3 UKOPA

4 PNGPC

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

MileageFailure frequencyFailure frequency trend line

05101520253035

001020304050607

Figure 1 Mileage and failure frequency for oil pipelines

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

MileageFailure frequencyFailure frequency trend line

01020304050607080

0002004006008

01012014016

Figure 2 Mileage and failure frequency for gas pipelines

212 Failure Causes Based on the statistical results from2010 to 2015 which includes 432 oil pipeline failures and238 gas pipeline failures all of which are flagged as sig-nificant incidents in the database the top 3 causes for oilpipeline failures are corrosion pipeweldmaterial failure andequipment failure while those of gas pipeline failures arepipeweld material failure excavation damag and corrosion(see Figures 3 and 4)

Table 2 listed the causes and subcauses categorized byPHMSA

(1) Corrosion For liquid pipelines corrosion is the mostimportant factor for failure while for gas pipelines corrosionis the top 3 of all failure factors Among those externalcorrosion usually accounted for more than 60 mainly

10725

9622

7517

6415

266

256

246

153

CorrosionPipeweld material failureEquipment failureExcavation damage

Other outside force damageIncorrect operationOthersNatural force damage

Figure 3 Oil pipelines causes

5624

5222

4318

2711

209

208

156

52

Pipeweld material failureExcavation damageCorrosionEquipment failure

Natural force damageOther outside force damageOthersIncorrect operation

Figure 4 Gas pipelines causes

Journal of Petroleum Engineering 3

Table 2 Causes and sub-cause categried by PHMSA

NO Causes Sub-causes

1

Corrosion

External Corrosion Galvanic Corrosion Stray Corrosion Microbiological Corrosion SelectiveSeam Corrosion

Internal Corrosion Corrosive Commodity Acid Water Microbiological Corrosion Erosion

2

PipeWeld Material FailureConstruction Installation or Fabrication Related Weld Quality Mechanical Damage in the Field Original Manufacturing Related Weld Quality Manufacturing Defect Environmental Related Stress Corrosion Cracking Deformation Related Cracking

3

Excavation Damage

Operatorrsquos Contractor (Second Party) Excavation Practices not Sufficient Locating Practices not SufficientPrevious Damage

Third PartyExcavation Practices not Sufficient Locating Practices not SufficientOne-call Notification Practices not Sufficient One-call Notification CenterError

Previous Damage due to Excavation Activity One-call Notification Practices not Sufficient Previous Damage 4 Natural Force Damage Earth Movement Heavy RainsFloods Lighting Temperature

5 Incorrect Operation Damage by Operator or Operatorrsquos Contractor Pipeline or EquipmentOverpressure Equipment not Installed Properly

6 Other Outside Force Damage Damage by Cars Boats Nearby Industry or FireExplosion

galvanic corrosion while internal corrosion is mainly ofmicrobiological corrosion

(2) PipeWeld Material Failure If the failure incidents for oiland gas pipelines are analyzed together pipeweld materialfailure is the top 1 factor For subcauses that are analyzed con-struction related (including field welded girth weld backfilldent etc) accounts for more than 50

(3) ExcavationDamage Excavation damage is another impor-tant cause for oil and gas pipeline failures in the US whichaccounts for 15 for oil pipeline failures and 22 for naturalgas pipeline failures Among those the third party excavationdamage accounts for the largest percentage mainly due tothe usage of one-call (excavation call system) system and theinsufficient excavation practices

(4) Natural Force Damage In this cause earth movement andheavy rainflood are the main factors

213 Failure Consequences During 2004 to 2015 numbersof casualties and property loss caused by pipeline accidentsin the US did not vary significantly except the peak valuein 2010 (see Figure 5) which is because of the rupture fire ofthe Pacific Gas and Electric Companyrsquos pipeline and ruptureleakage of Enbridge 6B crude oil pipeline

The property loss includes estimated cost of publicand nonoperator private property damage product releasedintentionally or unintentionally operatorrsquos property damageand repairs operatorrsquos emergency response and environmen-tal remediation

22 EGIG Up to 2013 the total length of EGIG gas pipelinebecomes 144 kkm The objective of EGIG is to collect and

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Fatalities (oil amp gas)Injuries (oil amp gas)Property loss (oil amp gas) (million dollars)

010203040506070

020040060080010001200140016001800

Figure 5 Failure consequences in the US reported by PHMSA

present data on loss of gas incidents in order to presentthe safety performance of the European gas transmissionnetwork to the general public and authorities

The required criteria for an incident to be recorded in theEGIG database are the following

(1) The incidentmust lead to an unintentional gas release(2) The pipeline must fulfil the following conditions

(a) to be made of steel(b) to be onshore(c) to have a maximum operating pressure higher

than 15 bar(d) to be located outside the fences of the gas

installations

4 Journal of Petroleum Engineering

Year [mdash]

2012

2013

2009

2008

2007

2006

2005

2010

2011

2002

2001

2000

1999

1991

1990

1989

1987

1988

2003

2004

1985

1992

1993

1994

1995

1996

1997

1998

1984

1983

1982

1981

1980

1986

1978

1977

1976

1975

1974

1979

1973

1971

1972

1970

0

01

02

03

04

05

06

Failu

re fr

eque

ncy

per 1

000 k

mmiddoty

r

External interferenceConstruction defectmaterial failureGround movementCorrosionHot tap made by errorOther and unknown

Figure 6 Failure frequencies of different causes by EGIG

35

24

16

4

13

8

External interferenceCorrosionConstructionDefectsmaterial failures

Hot tapGround movementOtherunknown

Figure 7 Failure causes by EGIG

221 Failure Frequency From 1970 to 2013 the primaryfailure frequencies for the entire period (up to the year) percause keep decreasing (See Figure 6)

In 2013 the primary failure frequency over the entireperiod (1970ndash2013) was equal to 033 per kkmsdotyr This isslightly lower than the failure frequency of 035 per kkmsdotyrreported in the 8th EGIG report (1970ndash2010) The primaryfailure frequency over the last five years was equal to 016 perkkmsdotyr showing an improved performance over recent years

222 Failure Causes Top 3 causes for gas pipeline failures inEGIG are external interference corrosion and constructiondefectsmaterial failure (see Figure 7)

1 of 23

4 of 105 of 13

1 of 6

1 of 281 of 251 of 25

2 of 3

1 of 2

2 of 61 of 3

1 of 11 of 1

1 of 2

0102030405060708090

100

Perc

enta

ge o

f rup

ture

s tha

t ign

ited

()

Diameter (inches)

dlt

5㰀㰀

5㰀㰀

ledlt

11㰀㰀

11㰀㰀

ledlt

17㰀㰀

17㰀㰀

ledlt

23㰀㰀

23㰀㰀

ledlt

29㰀㰀

29㰀㰀

ledlt

35㰀㰀

35㰀㰀

ledlt

41㰀㰀

41㰀㰀

led

p le 35 bar35 bar lt p le 55 barp gt 55 bar

Figure 8 Ignited failures analysis by EGIG

00008

00061

00015

00023

00008

00046

00008

00015

Employeescontractors

Causing Fighting Public

InjuriesFatalities

0

0001

0002

0003

0004

0005

0006

0007

Perc

enta

ge o

f inc

iden

t (

)

Figure 9 Casualtiesrsquo analysis by EGIG

223 Failure Consequences According to statistical resultsin period of 1970ndash2013 only 50 of the gas releasesrecorded in the EGIG database ignited Gas releases fromlarge diameter pipelines at high pressure have ignited morefrequently than smaller diameter pipelines at lower pressure(see Figure 8)

The highest fatality and injury rate can be found amongthe people who are directly involved in causing the incidentEight cases (061 total 1309) caused fatalities among thepeople causing the incident (see Figure 9)

23 UKOPA Up to 2014 the total length of UKOPA pipelinebecomes 224 kkm A product loss incident is defined in thecontext of this report as

Journal of Petroleum Engineering 5

Overall average up to yearMoving 5-year average

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

Year [mdash]

0

02

04

06

08

1

12

14

Freq

uenc

y pe

r 100

0 kmmiddoty

r

Figure 10 Average failure frequencies by UKOPA

1962ndash20142010ndash2014

Cause

Exte

rnal

corr

osio

n

Exte

rnal

inte

rfere

nce

Gro

und

mov

emen

t

Inte

rnal

corr

osio

n

Girt

h w

eld

defe

ct

Oth

er

Pipe

def

ect

Seam

wel

d de

fect

Unk

now

n

0

001

002

003

004

005

006

Freq

uenc

y pe

r 100

0 kmmiddoty

r

Figure 11 Failure causes by UKOPA

(1) an unintentional loss of product from the pipeline(2) within the public domain and outside the fences of

installations(3) excluding associated equipment (eg valves com-

pressors) or parts other than the pipeline itself

231 Failure Frequency From 1962 to 2014 altogether 192leakages have been recorded The overall failure frequencyover the period 1962 to 2014 is 0219 incidents per kkmsdotyrwhile in the previous report this figurewas 0223 incidents perkkmsdotyr (covering the period from 1962 to 2013) The overalltrend continues to show a reduction in failure frequency (seeFigure 10)

232 Failure Causes The top 3 failure causes of UKOPAare external corrosion external interference and girth welddefects (see Figure 11)

Table 3 Ignited failures analysis by UKOPA

AffectedComponent Cause Of Fault Hole Diameter Class

Pipe SeamWeld Defect 0ndash6mm

Pipe Ground Movement Full Bore and Above(1810158401015840 Diameter pipe)

Pipe Girth Weld Defect 6ndash20mmPipe Unknown 6ndash20mmPipe Pipe Defect 0ndash6mmPipe Unknown 40ndash110mmPipe Lightning Strike 0ndash6mmBend Internal Corrosion 0ndash6mmBend Pipe Defect 6ndash20mm

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

Mileage (km)Leakage number (times)

0

5

10

15

20

25

0

10000

20000

30000

40000

50000

60000

Figure 12 Mileage and failure numbers for PNGPC pipelines

233 Failure Consequences There were 9 out of 192 (47)product loss incidents that resulted in ignition Howeverthere is no obvious conclusion as shown in Table 3

24 PNGPC Up to 2015 the total length of PNGPC long-distance pipeline becomes 53 kkm Failure data were filteredin order to be comparable with other countries Only unin-tentional leakages for crude oil refined oil and natural gasfor transmission lines are counted here

241 Failure Frequency Failure frequency has increasedbefore 2011 and decreased in recent 5 years (see Figures 12 and13)

242 Failure Causes During 2006 to 2015 altogether 134leakages have been recorded among which illegal tapmanufacturing defects and construction quality are the top3 causes (see Figure 14)

(1) Illegal Tap With the enhancement of legislation failure ofthis cause has a significant downward trend

(2) Manufacturing Defect The majority of failures belong-ing to this cause are due to spiral weld defects which is

6 Journal of Petroleum Engineering

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

Oil pipelineGas pipelineOil amp gas pipeline

0010203040506070809

1

Figure 13 Average failure frequencies by PNGPC

50

19

10

9

74 1

Illegal tapManufacturing defectConstruction qualityCorrosion

Incorrect operation

Third partyNatural force damage

Figure 14 Failure causes by PNGPC

significantly decreasing as the manufacturing quality im-proved and the vintage pipelines abandoned

(3) Construction Quality This mainly includes girth welddefects gouges and dents As the decreasing of illegaltap and manufacturing defects and also the well-controlledcorrosion construction quality becomes the top concernedfactor in China now

(4) Corrosion As for the abandonment of old pipelinesand all kinds of control methods including periodic in-lineinspection corrosion is not as serious as before Howeverpin-hole corrosions still cause failures which are hardlydetected by MFL

(5) Third Party This is mainly due to the supervision escap-ing with unauthorized construction of third party which

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

PHSMAPNGPC

0010203040506070809

1

Figure 15 Compared failure frequencies for oil pipelines

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

0002004006008

01012014016018

02

PHSMAPNGPC

UKOPAEGIG

Figure 16 Compared failure frequencies for gas pipelines

causes damage on the pipes As third party constructionsmostly happen in the economically active regions conse-quences are relatively serious

3 Analysis Result Comparisons

Investigation and referring to foreign related failure statisticscan provide good experience for domestic pipeline operatorswhile figuring out their own management level for continu-ous improvement

31 Failure Frequencies Compared with the failure frequen-cies (5 yearsrsquo moving average) at home and abroad in the past10 years the value of PNGPCrsquos oil pipelines is higher than thatof theUS while that of PNGPCrsquos gas pipeline is roughly at thesame level compared to the US and slightly lower than theEuropean (see Figures 15 and 16)

Journal of Petroleum Engineering 7

32 Failure Causes

(1) Illegal Tap Restricted by the current situation of Chinarsquossocial and economic development this cause becomesChinarsquos top 1 factor of pipeline failure during the last 10 yearswhich rarely happens in the US and Europe However withthe strengthening of legislation and publicity the frequencyis significantly reduced

(2) PipeWeld Material Failure (Manufacturing Defect andConstruction Defect) This factor leads to a higher proportionof failures both at home and abroad With the promotion oftechnology and quality management defects on pipe bodiescan be effectively controlled However due to various con-straints it is difficult to guarantee the construction qualityHence girth weld defects dents gouges and other defectswill still exist to certain degrees

(3) Corrosion As for the abandonment of vintage pipelinesand the promotion of integrity management corrosion fail-ures are effectively controlled which is still one of the mainfactors of failures generally

(4) Third Party Failures due to this factor always occur ineconomically active areas between urban and rural bothat home and abroad As for pipeline patrolling in the USand Europe is not as timely and intensive as in China theproportion of failure caused by this factor is slightly higherthan that of China

33 Failure Consequences Compared with foreign countriesfailure consequence data is quite deficient in China Only theconsequences of serious pipeline accidents will be recordedincluding casualties economic losses and leakage

4 Conclusions and Suggestions

As for the concern from government and public on pipelinesafety in Chinamanagement has been significantly improvedby operators Consequently frequency of pipeline failures isdecreasing According to the statistical results PNGPC is notvery far compared with foreign countriesThere are still someaspects both in technology and inmanagement that should beimproved such as quality ofmanufacture and construction ofpipeline and third party monitoring

Nomenclature

PHMSA Pipeline and Hazardous Materials SafetyAdministration

EGIG European Gas Pipeline Incident DataGroup

UKOPA United Kingdom Onshore PipelineOperatorsrsquo Association

PNGPC PetroChina Natural Gas amp PipelineCompany

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

Preparation of this paper was supported by the PNGPC thisis gratefully acknowledged by coauthors DongpoWang TingWang Qingshan Feng and Xinqi Yang Thanks are also dueto numerous past and present colleagues for insights andhelpful discussions

References

[1] P Hopkins ldquoTransmission pipelines how to improve theirintegrity and prevent failuresrdquo in Pipeline Technology Proceed-ings of the 2nd international pipeline technology conference RDenys Ed vol 1 pp 683ndash702 1995

[2] W Y Zheng ldquoStress corrosion cracking of oil and gas pipelinesin near neutral pH environment Review of recent researchrdquoEnergy Materials Materials Science and Engineering for EnergySystems vol 3 no 4 pp 220ndash226 2008

[3] T Wang W Xuan X Wang et al ldquoOverview of oil and gaspipeline failure databaserdquo in Proceedings of International Con-ference on Pipelines and Trenchless Technology ICPTT 2013 pp1161ndash1167 chn October 2013

[4] H L Li X W Zhao and L K Ji ldquoOil and gas pipeline failureanalysis and integrity managementrdquo Testing - Physical Volumepp 24ndash31 2005

[5] R Singh Pipeline Integrity Handbook Risk Management andEvaluation vol 51 Gulf Professional Publishing UK 2014

[6] C R F Azevedo ldquoFailure analysis of a crude oil pipelinerdquo Engi-neering Failure Analysis vol 14 no 6 pp 978ndash994 2007

[7] PHMSA httpprimisphmsadotgovcommreportssafetyPSIhtml 2015

[8] EGIG ldquo9th report of the gas pipeline incidents of european gaspipeline incident data grouprdquo 2015

[9] UKOPA ldquoUKOPA pipeline product loss incidents and faultsreport (1962ndash2014)rdquo 2015

Journal of Petroleum Engineering 3

Table 2 Causes and sub-cause categried by PHMSA

NO Causes Sub-causes

1

Corrosion

External Corrosion Galvanic Corrosion Stray Corrosion Microbiological Corrosion SelectiveSeam Corrosion

Internal Corrosion Corrosive Commodity Acid Water Microbiological Corrosion Erosion

2

PipeWeld Material FailureConstruction Installation or Fabrication Related Weld Quality Mechanical Damage in the Field Original Manufacturing Related Weld Quality Manufacturing Defect Environmental Related Stress Corrosion Cracking Deformation Related Cracking

3

Excavation Damage

Operatorrsquos Contractor (Second Party) Excavation Practices not Sufficient Locating Practices not SufficientPrevious Damage

Third PartyExcavation Practices not Sufficient Locating Practices not SufficientOne-call Notification Practices not Sufficient One-call Notification CenterError

Previous Damage due to Excavation Activity One-call Notification Practices not Sufficient Previous Damage 4 Natural Force Damage Earth Movement Heavy RainsFloods Lighting Temperature

5 Incorrect Operation Damage by Operator or Operatorrsquos Contractor Pipeline or EquipmentOverpressure Equipment not Installed Properly

6 Other Outside Force Damage Damage by Cars Boats Nearby Industry or FireExplosion

galvanic corrosion while internal corrosion is mainly ofmicrobiological corrosion

(2) PipeWeld Material Failure If the failure incidents for oiland gas pipelines are analyzed together pipeweld materialfailure is the top 1 factor For subcauses that are analyzed con-struction related (including field welded girth weld backfilldent etc) accounts for more than 50

(3) ExcavationDamage Excavation damage is another impor-tant cause for oil and gas pipeline failures in the US whichaccounts for 15 for oil pipeline failures and 22 for naturalgas pipeline failures Among those the third party excavationdamage accounts for the largest percentage mainly due tothe usage of one-call (excavation call system) system and theinsufficient excavation practices

(4) Natural Force Damage In this cause earth movement andheavy rainflood are the main factors

213 Failure Consequences During 2004 to 2015 numbersof casualties and property loss caused by pipeline accidentsin the US did not vary significantly except the peak valuein 2010 (see Figure 5) which is because of the rupture fire ofthe Pacific Gas and Electric Companyrsquos pipeline and ruptureleakage of Enbridge 6B crude oil pipeline

The property loss includes estimated cost of publicand nonoperator private property damage product releasedintentionally or unintentionally operatorrsquos property damageand repairs operatorrsquos emergency response and environmen-tal remediation

22 EGIG Up to 2013 the total length of EGIG gas pipelinebecomes 144 kkm The objective of EGIG is to collect and

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

Fatalities (oil amp gas)Injuries (oil amp gas)Property loss (oil amp gas) (million dollars)

010203040506070

020040060080010001200140016001800

Figure 5 Failure consequences in the US reported by PHMSA

present data on loss of gas incidents in order to presentthe safety performance of the European gas transmissionnetwork to the general public and authorities

The required criteria for an incident to be recorded in theEGIG database are the following

(1) The incidentmust lead to an unintentional gas release(2) The pipeline must fulfil the following conditions

(a) to be made of steel(b) to be onshore(c) to have a maximum operating pressure higher

than 15 bar(d) to be located outside the fences of the gas

installations

4 Journal of Petroleum Engineering

Year [mdash]

2012

2013

2009

2008

2007

2006

2005

2010

2011

2002

2001

2000

1999

1991

1990

1989

1987

1988

2003

2004

1985

1992

1993

1994

1995

1996

1997

1998

1984

1983

1982

1981

1980

1986

1978

1977

1976

1975

1974

1979

1973

1971

1972

1970

0

01

02

03

04

05

06

Failu

re fr

eque

ncy

per 1

000 k

mmiddoty

r

External interferenceConstruction defectmaterial failureGround movementCorrosionHot tap made by errorOther and unknown

Figure 6 Failure frequencies of different causes by EGIG

35

24

16

4

13

8

External interferenceCorrosionConstructionDefectsmaterial failures

Hot tapGround movementOtherunknown

Figure 7 Failure causes by EGIG

221 Failure Frequency From 1970 to 2013 the primaryfailure frequencies for the entire period (up to the year) percause keep decreasing (See Figure 6)

In 2013 the primary failure frequency over the entireperiod (1970ndash2013) was equal to 033 per kkmsdotyr This isslightly lower than the failure frequency of 035 per kkmsdotyrreported in the 8th EGIG report (1970ndash2010) The primaryfailure frequency over the last five years was equal to 016 perkkmsdotyr showing an improved performance over recent years

222 Failure Causes Top 3 causes for gas pipeline failures inEGIG are external interference corrosion and constructiondefectsmaterial failure (see Figure 7)

1 of 23

4 of 105 of 13

1 of 6

1 of 281 of 251 of 25

2 of 3

1 of 2

2 of 61 of 3

1 of 11 of 1

1 of 2

0102030405060708090

100

Perc

enta

ge o

f rup

ture

s tha

t ign

ited

()

Diameter (inches)

dlt

5㰀㰀

5㰀㰀

ledlt

11㰀㰀

11㰀㰀

ledlt

17㰀㰀

17㰀㰀

ledlt

23㰀㰀

23㰀㰀

ledlt

29㰀㰀

29㰀㰀

ledlt

35㰀㰀

35㰀㰀

ledlt

41㰀㰀

41㰀㰀

led

p le 35 bar35 bar lt p le 55 barp gt 55 bar

Figure 8 Ignited failures analysis by EGIG

00008

00061

00015

00023

00008

00046

00008

00015

Employeescontractors

Causing Fighting Public

InjuriesFatalities

0

0001

0002

0003

0004

0005

0006

0007

Perc

enta

ge o

f inc

iden

t (

)

Figure 9 Casualtiesrsquo analysis by EGIG

223 Failure Consequences According to statistical resultsin period of 1970ndash2013 only 50 of the gas releasesrecorded in the EGIG database ignited Gas releases fromlarge diameter pipelines at high pressure have ignited morefrequently than smaller diameter pipelines at lower pressure(see Figure 8)

The highest fatality and injury rate can be found amongthe people who are directly involved in causing the incidentEight cases (061 total 1309) caused fatalities among thepeople causing the incident (see Figure 9)

23 UKOPA Up to 2014 the total length of UKOPA pipelinebecomes 224 kkm A product loss incident is defined in thecontext of this report as

Journal of Petroleum Engineering 5

Overall average up to yearMoving 5-year average

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

Year [mdash]

0

02

04

06

08

1

12

14

Freq

uenc

y pe

r 100

0 kmmiddoty

r

Figure 10 Average failure frequencies by UKOPA

1962ndash20142010ndash2014

Cause

Exte

rnal

corr

osio

n

Exte

rnal

inte

rfere

nce

Gro

und

mov

emen

t

Inte

rnal

corr

osio

n

Girt

h w

eld

defe

ct

Oth

er

Pipe

def

ect

Seam

wel

d de

fect

Unk

now

n

0

001

002

003

004

005

006

Freq

uenc

y pe

r 100

0 kmmiddoty

r

Figure 11 Failure causes by UKOPA

(1) an unintentional loss of product from the pipeline(2) within the public domain and outside the fences of

installations(3) excluding associated equipment (eg valves com-

pressors) or parts other than the pipeline itself

231 Failure Frequency From 1962 to 2014 altogether 192leakages have been recorded The overall failure frequencyover the period 1962 to 2014 is 0219 incidents per kkmsdotyrwhile in the previous report this figurewas 0223 incidents perkkmsdotyr (covering the period from 1962 to 2013) The overalltrend continues to show a reduction in failure frequency (seeFigure 10)

232 Failure Causes The top 3 failure causes of UKOPAare external corrosion external interference and girth welddefects (see Figure 11)

Table 3 Ignited failures analysis by UKOPA

AffectedComponent Cause Of Fault Hole Diameter Class

Pipe SeamWeld Defect 0ndash6mm

Pipe Ground Movement Full Bore and Above(1810158401015840 Diameter pipe)

Pipe Girth Weld Defect 6ndash20mmPipe Unknown 6ndash20mmPipe Pipe Defect 0ndash6mmPipe Unknown 40ndash110mmPipe Lightning Strike 0ndash6mmBend Internal Corrosion 0ndash6mmBend Pipe Defect 6ndash20mm

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

Mileage (km)Leakage number (times)

0

5

10

15

20

25

0

10000

20000

30000

40000

50000

60000

Figure 12 Mileage and failure numbers for PNGPC pipelines

233 Failure Consequences There were 9 out of 192 (47)product loss incidents that resulted in ignition Howeverthere is no obvious conclusion as shown in Table 3

24 PNGPC Up to 2015 the total length of PNGPC long-distance pipeline becomes 53 kkm Failure data were filteredin order to be comparable with other countries Only unin-tentional leakages for crude oil refined oil and natural gasfor transmission lines are counted here

241 Failure Frequency Failure frequency has increasedbefore 2011 and decreased in recent 5 years (see Figures 12 and13)

242 Failure Causes During 2006 to 2015 altogether 134leakages have been recorded among which illegal tapmanufacturing defects and construction quality are the top3 causes (see Figure 14)

(1) Illegal Tap With the enhancement of legislation failure ofthis cause has a significant downward trend

(2) Manufacturing Defect The majority of failures belong-ing to this cause are due to spiral weld defects which is

6 Journal of Petroleum Engineering

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

Oil pipelineGas pipelineOil amp gas pipeline

0010203040506070809

1

Figure 13 Average failure frequencies by PNGPC

50

19

10

9

74 1

Illegal tapManufacturing defectConstruction qualityCorrosion

Incorrect operation

Third partyNatural force damage

Figure 14 Failure causes by PNGPC

significantly decreasing as the manufacturing quality im-proved and the vintage pipelines abandoned

(3) Construction Quality This mainly includes girth welddefects gouges and dents As the decreasing of illegaltap and manufacturing defects and also the well-controlledcorrosion construction quality becomes the top concernedfactor in China now

(4) Corrosion As for the abandonment of old pipelinesand all kinds of control methods including periodic in-lineinspection corrosion is not as serious as before Howeverpin-hole corrosions still cause failures which are hardlydetected by MFL

(5) Third Party This is mainly due to the supervision escap-ing with unauthorized construction of third party which

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

PHSMAPNGPC

0010203040506070809

1

Figure 15 Compared failure frequencies for oil pipelines

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

0002004006008

01012014016018

02

PHSMAPNGPC

UKOPAEGIG

Figure 16 Compared failure frequencies for gas pipelines

causes damage on the pipes As third party constructionsmostly happen in the economically active regions conse-quences are relatively serious

3 Analysis Result Comparisons

Investigation and referring to foreign related failure statisticscan provide good experience for domestic pipeline operatorswhile figuring out their own management level for continu-ous improvement

31 Failure Frequencies Compared with the failure frequen-cies (5 yearsrsquo moving average) at home and abroad in the past10 years the value of PNGPCrsquos oil pipelines is higher than thatof theUS while that of PNGPCrsquos gas pipeline is roughly at thesame level compared to the US and slightly lower than theEuropean (see Figures 15 and 16)

Journal of Petroleum Engineering 7

32 Failure Causes

(1) Illegal Tap Restricted by the current situation of Chinarsquossocial and economic development this cause becomesChinarsquos top 1 factor of pipeline failure during the last 10 yearswhich rarely happens in the US and Europe However withthe strengthening of legislation and publicity the frequencyis significantly reduced

(2) PipeWeld Material Failure (Manufacturing Defect andConstruction Defect) This factor leads to a higher proportionof failures both at home and abroad With the promotion oftechnology and quality management defects on pipe bodiescan be effectively controlled However due to various con-straints it is difficult to guarantee the construction qualityHence girth weld defects dents gouges and other defectswill still exist to certain degrees

(3) Corrosion As for the abandonment of vintage pipelinesand the promotion of integrity management corrosion fail-ures are effectively controlled which is still one of the mainfactors of failures generally

(4) Third Party Failures due to this factor always occur ineconomically active areas between urban and rural bothat home and abroad As for pipeline patrolling in the USand Europe is not as timely and intensive as in China theproportion of failure caused by this factor is slightly higherthan that of China

33 Failure Consequences Compared with foreign countriesfailure consequence data is quite deficient in China Only theconsequences of serious pipeline accidents will be recordedincluding casualties economic losses and leakage

4 Conclusions and Suggestions

As for the concern from government and public on pipelinesafety in Chinamanagement has been significantly improvedby operators Consequently frequency of pipeline failures isdecreasing According to the statistical results PNGPC is notvery far compared with foreign countriesThere are still someaspects both in technology and inmanagement that should beimproved such as quality ofmanufacture and construction ofpipeline and third party monitoring

Nomenclature

PHMSA Pipeline and Hazardous Materials SafetyAdministration

EGIG European Gas Pipeline Incident DataGroup

UKOPA United Kingdom Onshore PipelineOperatorsrsquo Association

PNGPC PetroChina Natural Gas amp PipelineCompany

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

Preparation of this paper was supported by the PNGPC thisis gratefully acknowledged by coauthors DongpoWang TingWang Qingshan Feng and Xinqi Yang Thanks are also dueto numerous past and present colleagues for insights andhelpful discussions

References

[1] P Hopkins ldquoTransmission pipelines how to improve theirintegrity and prevent failuresrdquo in Pipeline Technology Proceed-ings of the 2nd international pipeline technology conference RDenys Ed vol 1 pp 683ndash702 1995

[2] W Y Zheng ldquoStress corrosion cracking of oil and gas pipelinesin near neutral pH environment Review of recent researchrdquoEnergy Materials Materials Science and Engineering for EnergySystems vol 3 no 4 pp 220ndash226 2008

[3] T Wang W Xuan X Wang et al ldquoOverview of oil and gaspipeline failure databaserdquo in Proceedings of International Con-ference on Pipelines and Trenchless Technology ICPTT 2013 pp1161ndash1167 chn October 2013

[4] H L Li X W Zhao and L K Ji ldquoOil and gas pipeline failureanalysis and integrity managementrdquo Testing - Physical Volumepp 24ndash31 2005

[5] R Singh Pipeline Integrity Handbook Risk Management andEvaluation vol 51 Gulf Professional Publishing UK 2014

[6] C R F Azevedo ldquoFailure analysis of a crude oil pipelinerdquo Engi-neering Failure Analysis vol 14 no 6 pp 978ndash994 2007

[7] PHMSA httpprimisphmsadotgovcommreportssafetyPSIhtml 2015

[8] EGIG ldquo9th report of the gas pipeline incidents of european gaspipeline incident data grouprdquo 2015

[9] UKOPA ldquoUKOPA pipeline product loss incidents and faultsreport (1962ndash2014)rdquo 2015

4 Journal of Petroleum Engineering

Year [mdash]

2012

2013

2009

2008

2007

2006

2005

2010

2011

2002

2001

2000

1999

1991

1990

1989

1987

1988

2003

2004

1985

1992

1993

1994

1995

1996

1997

1998

1984

1983

1982

1981

1980

1986

1978

1977

1976

1975

1974

1979

1973

1971

1972

1970

0

01

02

03

04

05

06

Failu

re fr

eque

ncy

per 1

000 k

mmiddoty

r

External interferenceConstruction defectmaterial failureGround movementCorrosionHot tap made by errorOther and unknown

Figure 6 Failure frequencies of different causes by EGIG

35

24

16

4

13

8

External interferenceCorrosionConstructionDefectsmaterial failures

Hot tapGround movementOtherunknown

Figure 7 Failure causes by EGIG

221 Failure Frequency From 1970 to 2013 the primaryfailure frequencies for the entire period (up to the year) percause keep decreasing (See Figure 6)

In 2013 the primary failure frequency over the entireperiod (1970ndash2013) was equal to 033 per kkmsdotyr This isslightly lower than the failure frequency of 035 per kkmsdotyrreported in the 8th EGIG report (1970ndash2010) The primaryfailure frequency over the last five years was equal to 016 perkkmsdotyr showing an improved performance over recent years

222 Failure Causes Top 3 causes for gas pipeline failures inEGIG are external interference corrosion and constructiondefectsmaterial failure (see Figure 7)

1 of 23

4 of 105 of 13

1 of 6

1 of 281 of 251 of 25

2 of 3

1 of 2

2 of 61 of 3

1 of 11 of 1

1 of 2

0102030405060708090

100

Perc

enta

ge o

f rup

ture

s tha

t ign

ited

()

Diameter (inches)

dlt

5㰀㰀

5㰀㰀

ledlt

11㰀㰀

11㰀㰀

ledlt

17㰀㰀

17㰀㰀

ledlt

23㰀㰀

23㰀㰀

ledlt

29㰀㰀

29㰀㰀

ledlt

35㰀㰀

35㰀㰀

ledlt

41㰀㰀

41㰀㰀

led

p le 35 bar35 bar lt p le 55 barp gt 55 bar

Figure 8 Ignited failures analysis by EGIG

00008

00061

00015

00023

00008

00046

00008

00015

Employeescontractors

Causing Fighting Public

InjuriesFatalities

0

0001

0002

0003

0004

0005

0006

0007

Perc

enta

ge o

f inc

iden

t (

)

Figure 9 Casualtiesrsquo analysis by EGIG

223 Failure Consequences According to statistical resultsin period of 1970ndash2013 only 50 of the gas releasesrecorded in the EGIG database ignited Gas releases fromlarge diameter pipelines at high pressure have ignited morefrequently than smaller diameter pipelines at lower pressure(see Figure 8)

The highest fatality and injury rate can be found amongthe people who are directly involved in causing the incidentEight cases (061 total 1309) caused fatalities among thepeople causing the incident (see Figure 9)

23 UKOPA Up to 2014 the total length of UKOPA pipelinebecomes 224 kkm A product loss incident is defined in thecontext of this report as

Journal of Petroleum Engineering 5

Overall average up to yearMoving 5-year average

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

Year [mdash]

0

02

04

06

08

1

12

14

Freq

uenc

y pe

r 100

0 kmmiddoty

r

Figure 10 Average failure frequencies by UKOPA

1962ndash20142010ndash2014

Cause

Exte

rnal

corr

osio

n

Exte

rnal

inte

rfere

nce

Gro

und

mov

emen

t

Inte

rnal

corr

osio

n

Girt

h w

eld

defe

ct

Oth

er

Pipe

def

ect

Seam

wel

d de

fect

Unk

now

n

0

001

002

003

004

005

006

Freq

uenc

y pe

r 100

0 kmmiddoty

r

Figure 11 Failure causes by UKOPA

(1) an unintentional loss of product from the pipeline(2) within the public domain and outside the fences of

installations(3) excluding associated equipment (eg valves com-

pressors) or parts other than the pipeline itself

231 Failure Frequency From 1962 to 2014 altogether 192leakages have been recorded The overall failure frequencyover the period 1962 to 2014 is 0219 incidents per kkmsdotyrwhile in the previous report this figurewas 0223 incidents perkkmsdotyr (covering the period from 1962 to 2013) The overalltrend continues to show a reduction in failure frequency (seeFigure 10)

232 Failure Causes The top 3 failure causes of UKOPAare external corrosion external interference and girth welddefects (see Figure 11)

Table 3 Ignited failures analysis by UKOPA

AffectedComponent Cause Of Fault Hole Diameter Class

Pipe SeamWeld Defect 0ndash6mm

Pipe Ground Movement Full Bore and Above(1810158401015840 Diameter pipe)

Pipe Girth Weld Defect 6ndash20mmPipe Unknown 6ndash20mmPipe Pipe Defect 0ndash6mmPipe Unknown 40ndash110mmPipe Lightning Strike 0ndash6mmBend Internal Corrosion 0ndash6mmBend Pipe Defect 6ndash20mm

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

Mileage (km)Leakage number (times)

0

5

10

15

20

25

0

10000

20000

30000

40000

50000

60000

Figure 12 Mileage and failure numbers for PNGPC pipelines

233 Failure Consequences There were 9 out of 192 (47)product loss incidents that resulted in ignition Howeverthere is no obvious conclusion as shown in Table 3

24 PNGPC Up to 2015 the total length of PNGPC long-distance pipeline becomes 53 kkm Failure data were filteredin order to be comparable with other countries Only unin-tentional leakages for crude oil refined oil and natural gasfor transmission lines are counted here

241 Failure Frequency Failure frequency has increasedbefore 2011 and decreased in recent 5 years (see Figures 12 and13)

242 Failure Causes During 2006 to 2015 altogether 134leakages have been recorded among which illegal tapmanufacturing defects and construction quality are the top3 causes (see Figure 14)

(1) Illegal Tap With the enhancement of legislation failure ofthis cause has a significant downward trend

(2) Manufacturing Defect The majority of failures belong-ing to this cause are due to spiral weld defects which is

6 Journal of Petroleum Engineering

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

Oil pipelineGas pipelineOil amp gas pipeline

0010203040506070809

1

Figure 13 Average failure frequencies by PNGPC

50

19

10

9

74 1

Illegal tapManufacturing defectConstruction qualityCorrosion

Incorrect operation

Third partyNatural force damage

Figure 14 Failure causes by PNGPC

significantly decreasing as the manufacturing quality im-proved and the vintage pipelines abandoned

(3) Construction Quality This mainly includes girth welddefects gouges and dents As the decreasing of illegaltap and manufacturing defects and also the well-controlledcorrosion construction quality becomes the top concernedfactor in China now

(4) Corrosion As for the abandonment of old pipelinesand all kinds of control methods including periodic in-lineinspection corrosion is not as serious as before Howeverpin-hole corrosions still cause failures which are hardlydetected by MFL

(5) Third Party This is mainly due to the supervision escap-ing with unauthorized construction of third party which

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

PHSMAPNGPC

0010203040506070809

1

Figure 15 Compared failure frequencies for oil pipelines

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

0002004006008

01012014016018

02

PHSMAPNGPC

UKOPAEGIG

Figure 16 Compared failure frequencies for gas pipelines

causes damage on the pipes As third party constructionsmostly happen in the economically active regions conse-quences are relatively serious

3 Analysis Result Comparisons

Investigation and referring to foreign related failure statisticscan provide good experience for domestic pipeline operatorswhile figuring out their own management level for continu-ous improvement

31 Failure Frequencies Compared with the failure frequen-cies (5 yearsrsquo moving average) at home and abroad in the past10 years the value of PNGPCrsquos oil pipelines is higher than thatof theUS while that of PNGPCrsquos gas pipeline is roughly at thesame level compared to the US and slightly lower than theEuropean (see Figures 15 and 16)

Journal of Petroleum Engineering 7

32 Failure Causes

(1) Illegal Tap Restricted by the current situation of Chinarsquossocial and economic development this cause becomesChinarsquos top 1 factor of pipeline failure during the last 10 yearswhich rarely happens in the US and Europe However withthe strengthening of legislation and publicity the frequencyis significantly reduced

(2) PipeWeld Material Failure (Manufacturing Defect andConstruction Defect) This factor leads to a higher proportionof failures both at home and abroad With the promotion oftechnology and quality management defects on pipe bodiescan be effectively controlled However due to various con-straints it is difficult to guarantee the construction qualityHence girth weld defects dents gouges and other defectswill still exist to certain degrees

(3) Corrosion As for the abandonment of vintage pipelinesand the promotion of integrity management corrosion fail-ures are effectively controlled which is still one of the mainfactors of failures generally

(4) Third Party Failures due to this factor always occur ineconomically active areas between urban and rural bothat home and abroad As for pipeline patrolling in the USand Europe is not as timely and intensive as in China theproportion of failure caused by this factor is slightly higherthan that of China

33 Failure Consequences Compared with foreign countriesfailure consequence data is quite deficient in China Only theconsequences of serious pipeline accidents will be recordedincluding casualties economic losses and leakage

4 Conclusions and Suggestions

As for the concern from government and public on pipelinesafety in Chinamanagement has been significantly improvedby operators Consequently frequency of pipeline failures isdecreasing According to the statistical results PNGPC is notvery far compared with foreign countriesThere are still someaspects both in technology and inmanagement that should beimproved such as quality ofmanufacture and construction ofpipeline and third party monitoring

Nomenclature

PHMSA Pipeline and Hazardous Materials SafetyAdministration

EGIG European Gas Pipeline Incident DataGroup

UKOPA United Kingdom Onshore PipelineOperatorsrsquo Association

PNGPC PetroChina Natural Gas amp PipelineCompany

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

Preparation of this paper was supported by the PNGPC thisis gratefully acknowledged by coauthors DongpoWang TingWang Qingshan Feng and Xinqi Yang Thanks are also dueto numerous past and present colleagues for insights andhelpful discussions

References

[1] P Hopkins ldquoTransmission pipelines how to improve theirintegrity and prevent failuresrdquo in Pipeline Technology Proceed-ings of the 2nd international pipeline technology conference RDenys Ed vol 1 pp 683ndash702 1995

[2] W Y Zheng ldquoStress corrosion cracking of oil and gas pipelinesin near neutral pH environment Review of recent researchrdquoEnergy Materials Materials Science and Engineering for EnergySystems vol 3 no 4 pp 220ndash226 2008

[3] T Wang W Xuan X Wang et al ldquoOverview of oil and gaspipeline failure databaserdquo in Proceedings of International Con-ference on Pipelines and Trenchless Technology ICPTT 2013 pp1161ndash1167 chn October 2013

[4] H L Li X W Zhao and L K Ji ldquoOil and gas pipeline failureanalysis and integrity managementrdquo Testing - Physical Volumepp 24ndash31 2005

[5] R Singh Pipeline Integrity Handbook Risk Management andEvaluation vol 51 Gulf Professional Publishing UK 2014

[6] C R F Azevedo ldquoFailure analysis of a crude oil pipelinerdquo Engi-neering Failure Analysis vol 14 no 6 pp 978ndash994 2007

[7] PHMSA httpprimisphmsadotgovcommreportssafetyPSIhtml 2015

[8] EGIG ldquo9th report of the gas pipeline incidents of european gaspipeline incident data grouprdquo 2015

[9] UKOPA ldquoUKOPA pipeline product loss incidents and faultsreport (1962ndash2014)rdquo 2015

Journal of Petroleum Engineering 5

Overall average up to yearMoving 5-year average

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

Year [mdash]

0

02

04

06

08

1

12

14

Freq

uenc

y pe

r 100

0 kmmiddoty

r

Figure 10 Average failure frequencies by UKOPA

1962ndash20142010ndash2014

Cause

Exte

rnal

corr

osio

n

Exte

rnal

inte

rfere

nce

Gro

und

mov

emen

t

Inte

rnal

corr

osio

n

Girt

h w

eld

defe

ct

Oth

er

Pipe

def

ect

Seam

wel

d de

fect

Unk

now

n

0

001

002

003

004

005

006

Freq

uenc

y pe

r 100

0 kmmiddoty

r

Figure 11 Failure causes by UKOPA

(1) an unintentional loss of product from the pipeline(2) within the public domain and outside the fences of

installations(3) excluding associated equipment (eg valves com-

pressors) or parts other than the pipeline itself

231 Failure Frequency From 1962 to 2014 altogether 192leakages have been recorded The overall failure frequencyover the period 1962 to 2014 is 0219 incidents per kkmsdotyrwhile in the previous report this figurewas 0223 incidents perkkmsdotyr (covering the period from 1962 to 2013) The overalltrend continues to show a reduction in failure frequency (seeFigure 10)

232 Failure Causes The top 3 failure causes of UKOPAare external corrosion external interference and girth welddefects (see Figure 11)

Table 3 Ignited failures analysis by UKOPA

AffectedComponent Cause Of Fault Hole Diameter Class

Pipe SeamWeld Defect 0ndash6mm

Pipe Ground Movement Full Bore and Above(1810158401015840 Diameter pipe)

Pipe Girth Weld Defect 6ndash20mmPipe Unknown 6ndash20mmPipe Pipe Defect 0ndash6mmPipe Unknown 40ndash110mmPipe Lightning Strike 0ndash6mmBend Internal Corrosion 0ndash6mmBend Pipe Defect 6ndash20mm

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

Mileage (km)Leakage number (times)

0

5

10

15

20

25

0

10000

20000

30000

40000

50000

60000

Figure 12 Mileage and failure numbers for PNGPC pipelines

233 Failure Consequences There were 9 out of 192 (47)product loss incidents that resulted in ignition Howeverthere is no obvious conclusion as shown in Table 3

24 PNGPC Up to 2015 the total length of PNGPC long-distance pipeline becomes 53 kkm Failure data were filteredin order to be comparable with other countries Only unin-tentional leakages for crude oil refined oil and natural gasfor transmission lines are counted here

241 Failure Frequency Failure frequency has increasedbefore 2011 and decreased in recent 5 years (see Figures 12 and13)

242 Failure Causes During 2006 to 2015 altogether 134leakages have been recorded among which illegal tapmanufacturing defects and construction quality are the top3 causes (see Figure 14)

(1) Illegal Tap With the enhancement of legislation failure ofthis cause has a significant downward trend

(2) Manufacturing Defect The majority of failures belong-ing to this cause are due to spiral weld defects which is

6 Journal of Petroleum Engineering

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

Oil pipelineGas pipelineOil amp gas pipeline

0010203040506070809

1

Figure 13 Average failure frequencies by PNGPC

50

19

10

9

74 1

Illegal tapManufacturing defectConstruction qualityCorrosion

Incorrect operation

Third partyNatural force damage

Figure 14 Failure causes by PNGPC

significantly decreasing as the manufacturing quality im-proved and the vintage pipelines abandoned

(3) Construction Quality This mainly includes girth welddefects gouges and dents As the decreasing of illegaltap and manufacturing defects and also the well-controlledcorrosion construction quality becomes the top concernedfactor in China now

(4) Corrosion As for the abandonment of old pipelinesand all kinds of control methods including periodic in-lineinspection corrosion is not as serious as before Howeverpin-hole corrosions still cause failures which are hardlydetected by MFL

(5) Third Party This is mainly due to the supervision escap-ing with unauthorized construction of third party which

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

PHSMAPNGPC

0010203040506070809

1

Figure 15 Compared failure frequencies for oil pipelines

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

0002004006008

01012014016018

02

PHSMAPNGPC

UKOPAEGIG

Figure 16 Compared failure frequencies for gas pipelines

causes damage on the pipes As third party constructionsmostly happen in the economically active regions conse-quences are relatively serious

3 Analysis Result Comparisons

Investigation and referring to foreign related failure statisticscan provide good experience for domestic pipeline operatorswhile figuring out their own management level for continu-ous improvement

31 Failure Frequencies Compared with the failure frequen-cies (5 yearsrsquo moving average) at home and abroad in the past10 years the value of PNGPCrsquos oil pipelines is higher than thatof theUS while that of PNGPCrsquos gas pipeline is roughly at thesame level compared to the US and slightly lower than theEuropean (see Figures 15 and 16)

Journal of Petroleum Engineering 7

32 Failure Causes

(1) Illegal Tap Restricted by the current situation of Chinarsquossocial and economic development this cause becomesChinarsquos top 1 factor of pipeline failure during the last 10 yearswhich rarely happens in the US and Europe However withthe strengthening of legislation and publicity the frequencyis significantly reduced

(2) PipeWeld Material Failure (Manufacturing Defect andConstruction Defect) This factor leads to a higher proportionof failures both at home and abroad With the promotion oftechnology and quality management defects on pipe bodiescan be effectively controlled However due to various con-straints it is difficult to guarantee the construction qualityHence girth weld defects dents gouges and other defectswill still exist to certain degrees

(3) Corrosion As for the abandonment of vintage pipelinesand the promotion of integrity management corrosion fail-ures are effectively controlled which is still one of the mainfactors of failures generally

(4) Third Party Failures due to this factor always occur ineconomically active areas between urban and rural bothat home and abroad As for pipeline patrolling in the USand Europe is not as timely and intensive as in China theproportion of failure caused by this factor is slightly higherthan that of China

33 Failure Consequences Compared with foreign countriesfailure consequence data is quite deficient in China Only theconsequences of serious pipeline accidents will be recordedincluding casualties economic losses and leakage

4 Conclusions and Suggestions

As for the concern from government and public on pipelinesafety in Chinamanagement has been significantly improvedby operators Consequently frequency of pipeline failures isdecreasing According to the statistical results PNGPC is notvery far compared with foreign countriesThere are still someaspects both in technology and inmanagement that should beimproved such as quality ofmanufacture and construction ofpipeline and third party monitoring

Nomenclature

PHMSA Pipeline and Hazardous Materials SafetyAdministration

EGIG European Gas Pipeline Incident DataGroup

UKOPA United Kingdom Onshore PipelineOperatorsrsquo Association

PNGPC PetroChina Natural Gas amp PipelineCompany

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

Preparation of this paper was supported by the PNGPC thisis gratefully acknowledged by coauthors DongpoWang TingWang Qingshan Feng and Xinqi Yang Thanks are also dueto numerous past and present colleagues for insights andhelpful discussions

References

[1] P Hopkins ldquoTransmission pipelines how to improve theirintegrity and prevent failuresrdquo in Pipeline Technology Proceed-ings of the 2nd international pipeline technology conference RDenys Ed vol 1 pp 683ndash702 1995

[2] W Y Zheng ldquoStress corrosion cracking of oil and gas pipelinesin near neutral pH environment Review of recent researchrdquoEnergy Materials Materials Science and Engineering for EnergySystems vol 3 no 4 pp 220ndash226 2008

[3] T Wang W Xuan X Wang et al ldquoOverview of oil and gaspipeline failure databaserdquo in Proceedings of International Con-ference on Pipelines and Trenchless Technology ICPTT 2013 pp1161ndash1167 chn October 2013

[4] H L Li X W Zhao and L K Ji ldquoOil and gas pipeline failureanalysis and integrity managementrdquo Testing - Physical Volumepp 24ndash31 2005

[5] R Singh Pipeline Integrity Handbook Risk Management andEvaluation vol 51 Gulf Professional Publishing UK 2014

[6] C R F Azevedo ldquoFailure analysis of a crude oil pipelinerdquo Engi-neering Failure Analysis vol 14 no 6 pp 978ndash994 2007

[7] PHMSA httpprimisphmsadotgovcommreportssafetyPSIhtml 2015

[8] EGIG ldquo9th report of the gas pipeline incidents of european gaspipeline incident data grouprdquo 2015

[9] UKOPA ldquoUKOPA pipeline product loss incidents and faultsreport (1962ndash2014)rdquo 2015

6 Journal of Petroleum Engineering

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

Oil pipelineGas pipelineOil amp gas pipeline

0010203040506070809

1

Figure 13 Average failure frequencies by PNGPC

50

19

10

9

74 1

Illegal tapManufacturing defectConstruction qualityCorrosion

Incorrect operation

Third partyNatural force damage

Figure 14 Failure causes by PNGPC

significantly decreasing as the manufacturing quality im-proved and the vintage pipelines abandoned

(3) Construction Quality This mainly includes girth welddefects gouges and dents As the decreasing of illegaltap and manufacturing defects and also the well-controlledcorrosion construction quality becomes the top concernedfactor in China now

(4) Corrosion As for the abandonment of old pipelinesand all kinds of control methods including periodic in-lineinspection corrosion is not as serious as before Howeverpin-hole corrosions still cause failures which are hardlydetected by MFL

(5) Third Party This is mainly due to the supervision escap-ing with unauthorized construction of third party which

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

PHSMAPNGPC

0010203040506070809

1

Figure 15 Compared failure frequencies for oil pipelines

2006

ndash201

0

2007

ndash201

1

2008

ndash201

2

2009

ndash201

3

2010

ndash201

4

2011

ndash201

5

0002004006008

01012014016018

02

PHSMAPNGPC

UKOPAEGIG

Figure 16 Compared failure frequencies for gas pipelines

causes damage on the pipes As third party constructionsmostly happen in the economically active regions conse-quences are relatively serious

3 Analysis Result Comparisons

Investigation and referring to foreign related failure statisticscan provide good experience for domestic pipeline operatorswhile figuring out their own management level for continu-ous improvement

31 Failure Frequencies Compared with the failure frequen-cies (5 yearsrsquo moving average) at home and abroad in the past10 years the value of PNGPCrsquos oil pipelines is higher than thatof theUS while that of PNGPCrsquos gas pipeline is roughly at thesame level compared to the US and slightly lower than theEuropean (see Figures 15 and 16)

Journal of Petroleum Engineering 7

32 Failure Causes

(1) Illegal Tap Restricted by the current situation of Chinarsquossocial and economic development this cause becomesChinarsquos top 1 factor of pipeline failure during the last 10 yearswhich rarely happens in the US and Europe However withthe strengthening of legislation and publicity the frequencyis significantly reduced

(2) PipeWeld Material Failure (Manufacturing Defect andConstruction Defect) This factor leads to a higher proportionof failures both at home and abroad With the promotion oftechnology and quality management defects on pipe bodiescan be effectively controlled However due to various con-straints it is difficult to guarantee the construction qualityHence girth weld defects dents gouges and other defectswill still exist to certain degrees

(3) Corrosion As for the abandonment of vintage pipelinesand the promotion of integrity management corrosion fail-ures are effectively controlled which is still one of the mainfactors of failures generally

(4) Third Party Failures due to this factor always occur ineconomically active areas between urban and rural bothat home and abroad As for pipeline patrolling in the USand Europe is not as timely and intensive as in China theproportion of failure caused by this factor is slightly higherthan that of China

33 Failure Consequences Compared with foreign countriesfailure consequence data is quite deficient in China Only theconsequences of serious pipeline accidents will be recordedincluding casualties economic losses and leakage

4 Conclusions and Suggestions

As for the concern from government and public on pipelinesafety in Chinamanagement has been significantly improvedby operators Consequently frequency of pipeline failures isdecreasing According to the statistical results PNGPC is notvery far compared with foreign countriesThere are still someaspects both in technology and inmanagement that should beimproved such as quality ofmanufacture and construction ofpipeline and third party monitoring

Nomenclature

PHMSA Pipeline and Hazardous Materials SafetyAdministration

EGIG European Gas Pipeline Incident DataGroup

UKOPA United Kingdom Onshore PipelineOperatorsrsquo Association

PNGPC PetroChina Natural Gas amp PipelineCompany

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

Preparation of this paper was supported by the PNGPC thisis gratefully acknowledged by coauthors DongpoWang TingWang Qingshan Feng and Xinqi Yang Thanks are also dueto numerous past and present colleagues for insights andhelpful discussions

References

[1] P Hopkins ldquoTransmission pipelines how to improve theirintegrity and prevent failuresrdquo in Pipeline Technology Proceed-ings of the 2nd international pipeline technology conference RDenys Ed vol 1 pp 683ndash702 1995

[2] W Y Zheng ldquoStress corrosion cracking of oil and gas pipelinesin near neutral pH environment Review of recent researchrdquoEnergy Materials Materials Science and Engineering for EnergySystems vol 3 no 4 pp 220ndash226 2008

[3] T Wang W Xuan X Wang et al ldquoOverview of oil and gaspipeline failure databaserdquo in Proceedings of International Con-ference on Pipelines and Trenchless Technology ICPTT 2013 pp1161ndash1167 chn October 2013

[4] H L Li X W Zhao and L K Ji ldquoOil and gas pipeline failureanalysis and integrity managementrdquo Testing - Physical Volumepp 24ndash31 2005

[5] R Singh Pipeline Integrity Handbook Risk Management andEvaluation vol 51 Gulf Professional Publishing UK 2014

[6] C R F Azevedo ldquoFailure analysis of a crude oil pipelinerdquo Engi-neering Failure Analysis vol 14 no 6 pp 978ndash994 2007

[7] PHMSA httpprimisphmsadotgovcommreportssafetyPSIhtml 2015

[8] EGIG ldquo9th report of the gas pipeline incidents of european gaspipeline incident data grouprdquo 2015

[9] UKOPA ldquoUKOPA pipeline product loss incidents and faultsreport (1962ndash2014)rdquo 2015

Journal of Petroleum Engineering 7

32 Failure Causes

(1) Illegal Tap Restricted by the current situation of Chinarsquossocial and economic development this cause becomesChinarsquos top 1 factor of pipeline failure during the last 10 yearswhich rarely happens in the US and Europe However withthe strengthening of legislation and publicity the frequencyis significantly reduced

(2) PipeWeld Material Failure (Manufacturing Defect andConstruction Defect) This factor leads to a higher proportionof failures both at home and abroad With the promotion oftechnology and quality management defects on pipe bodiescan be effectively controlled However due to various con-straints it is difficult to guarantee the construction qualityHence girth weld defects dents gouges and other defectswill still exist to certain degrees

(3) Corrosion As for the abandonment of vintage pipelinesand the promotion of integrity management corrosion fail-ures are effectively controlled which is still one of the mainfactors of failures generally

(4) Third Party Failures due to this factor always occur ineconomically active areas between urban and rural bothat home and abroad As for pipeline patrolling in the USand Europe is not as timely and intensive as in China theproportion of failure caused by this factor is slightly higherthan that of China

33 Failure Consequences Compared with foreign countriesfailure consequence data is quite deficient in China Only theconsequences of serious pipeline accidents will be recordedincluding casualties economic losses and leakage

4 Conclusions and Suggestions

As for the concern from government and public on pipelinesafety in Chinamanagement has been significantly improvedby operators Consequently frequency of pipeline failures isdecreasing According to the statistical results PNGPC is notvery far compared with foreign countriesThere are still someaspects both in technology and inmanagement that should beimproved such as quality ofmanufacture and construction ofpipeline and third party monitoring

Nomenclature

PHMSA Pipeline and Hazardous Materials SafetyAdministration

EGIG European Gas Pipeline Incident DataGroup

UKOPA United Kingdom Onshore PipelineOperatorsrsquo Association

PNGPC PetroChina Natural Gas amp PipelineCompany

Conflicts of Interest

The authors declare that there are no conflicts of interestregarding the publication of this paper

Acknowledgments

Preparation of this paper was supported by the PNGPC thisis gratefully acknowledged by coauthors DongpoWang TingWang Qingshan Feng and Xinqi Yang Thanks are also dueto numerous past and present colleagues for insights andhelpful discussions

References

[1] P Hopkins ldquoTransmission pipelines how to improve theirintegrity and prevent failuresrdquo in Pipeline Technology Proceed-ings of the 2nd international pipeline technology conference RDenys Ed vol 1 pp 683ndash702 1995

[2] W Y Zheng ldquoStress corrosion cracking of oil and gas pipelinesin near neutral pH environment Review of recent researchrdquoEnergy Materials Materials Science and Engineering for EnergySystems vol 3 no 4 pp 220ndash226 2008

[3] T Wang W Xuan X Wang et al ldquoOverview of oil and gaspipeline failure databaserdquo in Proceedings of International Con-ference on Pipelines and Trenchless Technology ICPTT 2013 pp1161ndash1167 chn October 2013

[4] H L Li X W Zhao and L K Ji ldquoOil and gas pipeline failureanalysis and integrity managementrdquo Testing - Physical Volumepp 24ndash31 2005

[5] R Singh Pipeline Integrity Handbook Risk Management andEvaluation vol 51 Gulf Professional Publishing UK 2014

[6] C R F Azevedo ldquoFailure analysis of a crude oil pipelinerdquo Engi-neering Failure Analysis vol 14 no 6 pp 978ndash994 2007

[7] PHMSA httpprimisphmsadotgovcommreportssafetyPSIhtml 2015

[8] EGIG ldquo9th report of the gas pipeline incidents of european gaspipeline incident data grouprdquo 2015

[9] UKOPA ldquoUKOPA pipeline product loss incidents and faultsreport (1962ndash2014)rdquo 2015

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal of

Volume 201

Submit your manuscripts athttpswwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 201

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

DistributedSensor Networks

International Journal of