See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/268666715

Gas Detection for Offshore Application

Conference Paper · June 2014

CITATION READS

1 301

1 author:

Peter Okoh
Norwegian University of Science and Technology
21 PUBLICATIONS   133 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Peter Okoh on 20 May 2016.

The user has requested enhancement of the downloaded file.

 Gas Detection for Offshore Application
Peter Okoha
a
Norwegian University of Science and Technology, Trondheim, Norway

Abstract: Release of hazardous and flammable gas is a significant contributor to risk in the offshore oil
and gas industry and various types of automatic systems for rapid detection of gas are therefore installed
to accentuate the elimination or reduction of the dangerous releases. There are different types of gases
which may be released and gas may be released in different environments and under different conditions.
Several principles for detecting gas are therefore applied and a variety of types of gas detectors are in
use. However, a significant percentage of gas releases remain undetected by the dedicated detectors and
hence unaccounted for and uncontrolled.

The objectives of this paper are: (1) to present a state-of-the art overview of gas detection in relation to
offshore applications, (2) to present an overview of requirements for gas detection in the Norwegian off-
shore industry, and (3) to do a comparative study of performance standards for gas detection worldwide.
The paper builds on a review of literature, standards and guidelines in relation to gas detection offshore.

Keywords: Flammable, Toxic, Gas, Detection

1. INTRODUCTION

In the offshore industry, dangerous gases are naturally occurring or man-made in petroleum operations.
Release of hazardous and flammable gas is a significant contributor to risk in the oil and gas industry. The
ignition of flammable gas clouds or vapors can lead to major fire and explosion with highly devastating
consequences as was the case in the Piper Alpha disaster [7]. Similarly, toxic gas release can lead to
multiple fatalities over a wide area as was the case in the Bhopal gas tragedy [7], although this happened
in an onshore location.

Gas detection is a crucial topic in the process industries, e.g. the Norwegian offshore industry where
focus has been on safety barriers supported by the policy of reporting hydrocarbon leaks [20]. The
process industry learns from related incidents/accidents in addition to being proactive to predict what
can go wrong and how to control it. Various types of automatic systems for rapid detection of gas are
therefore installed to control the risk. In fact, the safety of the offshore industry depends on the efficiency
and effectiveness of the gas detection systems.

Gas detection has, however, experienced mixed results in the industry. Although it has achieved more
success than failure, the failure statistics is significant. According to the report of a research conducted
by [19], about 44% of all gas releases, or 38% of major gas releases were undetected by the gas detectors
deployed. The offshore environment is being characterized by a complex mix of open and enclosed
areas, low and high areas, diversity of hazardous gases and conditions for release, potential gas traps,
and varieties of operational and environmental conditions that may influence the unreliability of gas
detection. Besides, offshore installations have different challenging gas detection needs that require
specific solutions; e.g. some facilities require that detectors identify gases at the lowest possible level (in
ppm or LEL range), whereas other facilities are exposed to compounds or other gases that can undermine
the detectability of the target gases (i.e. a problem of cross-sensitivity or non-specificity). Furthermore,
complexities in ventilation patterns and in the size and composition of modules often make it extremely

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
difficult to site gas detectors; this implies that availability of detectors is no guarantee for detection [19].
Other gas detection related problems include incorrect selection of gas detectors, deficiencies in design,
installation, calibration and maintenance of gas detectors as well as users’ lack of knowledge of the
limitations of a given detection principle.

Several authors and companies have made or are making efforts to improve gas detection technologies.
Work on enhancing the effectiveness of single technology has been carried out by several authors, e.g.
[18] and [8] etc. The optimal placement of sensors under uncertainty has been studied by [12]. Fur-
thermore, the development of versatile single technology or integrated technologies for multiple gas
detection has been or is being explored by several others, e.g. [16], [5, 6] and [11].

The main objective of this paper is to present a state-of-the art overview on gas detection in relation to
offshore applications. The rest of the paper is structured as follows. First, the types of releases that may
occur and the conditions under which they may occur are described. This is followed by a description of
relevant principles for detecting gas, and relevant Norwegian regulatory requirements and standard. A
comparative study of gas detection standards worldwide is then presented, followed by conclusion and
recommendations.

2. TYPES OF GASEOUS RELEASES, THEIR CHARACTERISTICS AND POTEN-

TIAL SOURCES

The different types of gases which may be released in the offshore petroleum industry are described in
the following:

• Hydrocarbon (HC) gas release: This is a flammable release that may occur at atmospheric condi-
tions or under pressure from containment systems [13]. Such releases can occur in containment
systems subject to the following failure mechanisms: corrosion, erosion, wear, manufacturing de-
fects, operational loading, well pressure etc. Other possible causes include human factors in the
form of normal operational releases, operators error and third party damage [13]. The release can
occur at the topside where the process equipment are located or subsea in the form of blowouts
from wells and leaks from subsea pipelines and isolation valves etc. Specific areas where releases
are likely such as the rig floor, the vicinity of the test separator and the choke manifold require
permanent rather than portable gas detection system [9]. It is also common to release methane
(CH4 ), a lower hydrocarbon, from combustion to generate electricity and to power compressors
and pumps as well as from flaring of excess gas for safety and during well testing [15].
• Hydrogen Sulphide (H2 S ) release: This is an extremely toxic release that usually occurs as a con-
taminant in produced gases. It occurs naturally together with natural gases from wells. During
well testing, it is advisable to monitor the area to check the presence of Hydrogen Sulphide (H2 S )
concentrations and that it is safe for working, since even in relatively low concentrations this re-
lease can readily lead to fatality [9]. The first significant presence of H2 S is readily noticeable
from samples taken downstream of the choke manifold and at the gas outlet from the separator
[9]. Furthermore, H2 S usually collect at the lowest points on rigs such as the cellar deck area
(offshore) and on land rigs since it is heavier than air [9].
• Carbon Dioxide (CO2 ) release: This is a release that becomes dangerous usually in relatively high
levels in confined spaces. The release usually results from combustion of fossil fuels to generate
electricity and to power compressors and pumps, as well as from flaring of excess gas for safety
and during well testing [15].
• Carbon Monoxide (CO) release: This is a highly toxic release. It usually results from combustion
of fossil fuel to generate electricity and to power compressors and pumps, as well as from flaring

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
 of excess gas for safety [15].

3. PRINCIPLES FOR DETECTING GAS

Offshore gas detection system is necessary to warn about the presence of hazardous and flammable gases
in unacceptable concentrations within a given ambiance in order to prevent major accidents. Several
principles for detecting gas exist to cover the different types of gases which may be released under
different environments and conditions.

The types of gas detection technologies applicable to the offshore petroleum industry, a brief description
of their principles of operation as well as their safety-related applications are shown in Table 1.

Table 1: Operational principles and safety-related applications of gas detectors [1, 3, 10]

Operational Description of principles Applicable gases Safety-related

principles of gas applications
detectors
Catalytic Uses a catalytic bead to oxidize combustible gas; All combustible Flammable gas de-
a Wheatstone bridge converts the resulting change gases (non- tection
in resistance into a corresponding sensor signal. selectively)
Electrochemical Uses an electrochemical reaction to generate a Many toxic gases, Toxic gas detection
current proportional to the gas concentration. environmental pol-
lutants, combustion
products and oxygen.
Solid state Measures the change in resistance of a metal oxide HCs, CO, O3 , H2 S, Flammable gas de-
in response to the presence of a gas; the change in organic vapors, etc. tection
resistance translates into a concentration reading.
Thermal conduc- Measures the gas’ ability to transmit heat by com- Binary gas mixtures Flammable and toxic
tivity paring it with a reference gas (usually air). The (often a known gas in gas detection
change in electrical resistance as a result of the air); combustible and
heat transmission is proportional to the gas con- toxic gases
centration.
Photoacoustic In- Uses a gas ability to absorb IR radiation and gen- Many IR absorbing Flammable and toxic
frared (IR) erating an audible pressure pulse whose magni- gases; combustible gas detection
tude indicates the gas concentration present. gas, toxic gas
Infrared (Absorp- Applies absorption spectroscopy such that a spe- Many mid-IR ab- Flammable and toxic
tive) cific gas absorbs a specific wavelength in the in- sorbing gases, e.g. gas detection
frared (IR) spectrum, and the gas concentration is CO2 , CO, CH4 , NO
proportional to the amount of IR light absorbed. etc.
Ultrasonic (or Uses ultrasonic sensors to detect leak based on the All types of gases Flammable and toxic
acoustic) sound generated by escaping gas at ultrasonic fre- whether combustible, gas detection
quencies. toxic or inert.
IR gas cloud Applies an absorption imaging technique whereby Gases that absorb IR Flammable gas de-
imaging the image of an area illuminated by infrared radi- radiation at the wave- tection
ation is captured by an infrared camera. length of the IR radi-
ation, e.g. hydrocar-
bon gases.

The coverage of gas detection is a crucial factor to consider in addition to the vulnerabilities of the
detection technology that can be exploited by certain operational and environmental conditions. To this
end, the various principles for gas detection are classified based on coverage and each of the detection
principles was described further by its application area, strengths and weaknesses as shown in Table 2.

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
 Table 2: Further gas detection characteristics [10, 11, 14]

Coverage Detection Application areas Strengths Weaknesses

principles
Point detection Catalytic Point sources - poten- Robust, easily installable and operable, simple to calibrate, long Passive detection (not fail-safe), gas must diffuse into catalytic
tial leakage points (e.g. lifetime with a low life-cycle cost, detectability of a variety of bead so as to be detected, contaminants can poison or deacti-
pumps, compressors, gases, wide range of operating temperature, easily calibrated to vate catalyst, the only means of identifying loss of sensitivity
major packing, seal gases undetectable by infrared absorption, e.g. hydrogen. due to catalyst’s poison is by testing with appropriate gas reg-
or gasket vulnerable ularly, requires oxygen for detection, sensor performance may
points, etc). become degraded from prolonged exposure to high concentra-
tions of ignitable gas.
Electrochemical Same as above Speedy response, high accuracy, versatility (detects a wide Less effective at low ambient temperatures (≤ −40o C), cannot
range of toxic gases), low power consumption. withstand dry environment (<15% RH) over several months,
operates in a narrow pressure range (1±0.1 atm)
Solid state Same as above Robust, versatile (detects a wide range of gases), wide range of Usually not selective, although some new improvements have
operating temperature, resistant to corrosive and low-humidity overcome this limitation, high power consumption, operation is
environment, long operating life (2-10 years) not fail-safe.
Thermal con- Same as above Wide measuring range Non-specific (cross-sensitive), unsuitable for gases with ther-
ductivity mal conductivities (Tc) close to one. Gases with Tc < 1 are
more difficult to measure. Output signal not always linear.
Photoacoustic Same as above High sensitivity, linear output, simple to use, not subject to poi- Not suitable for hydrogen detection
Infrared (IR) soning, long-term stability
Fixed-point IR Same as above Immune to poisoning by contaminants, fail-safe operation, ab- Gas must pass by the sampling path so as to be detected, the
(absorptive) sence of routine calibration, can operate in the absence of oxy- gas to be detected must be infrared active (e.g. a hydrocar-
gen or in enriched oxygen, can operate in continuous presence bon), gases that do not absorb IR energy cannot be detected,
of gas highly humid and dusty environments can increase the mainte-
nance cost of IR detector, routine calibration to a different gas
is impractical, a relatively large amount of gas is required for
response testing, ambient temperature limit of detector use is
70o C, not suitable for multiple gas applications, the IR source
is not replaceable in the field, but in the factory.
Open path (line Open-path IR Boundaries with pub- Same as above and long line coverage. Same as above
or perimeter) (absorptive) lic areas and between
detection fire areas or equipment,
along rows of items and
perimeters

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
Area detection Ultrasonic (or General process areas, Very high detection rate of pressurized gas leaks, versatility (de- Unsuitable for low pressure leaks, under certain conditions in-
acoustic) loading/offloading tects pressurized leaks irrespective of gas type), unaffected by fluenced by artificial or natural ultrasonic sources, requires es-
facilities, gas turbines, ambient conditions (fog, heavy rain and others), minimal main- timation of background noise levels before installation, cannot
flow stations, tank tenance, absence of consumable parts, robust, fail-safe, insen- determine concentration of gas, cannot pin-point leak source.
farms etc. sitive to gas dilution and changing wind direction, wide area
coverage, gas must not be at the device for detection.
IR gas cloud Large gas clouds mon- Wide field of view and detection coverage, no gas calibration is Detectability of gases is poor when the contrast with the back-
imaging itoring in unmanned required in the field, highly immune to spurious alarm sources, ground is poor, heavy fog and rain reduces detection range, suit-
platforms, pipelines simultaneous detection of multiple gases. able only for large leaks - not a small leak detector.
4. REQUIREMENTS FROM NORWEGIAN REGULATIONS AND STANDARDS

Gas detection in the Norwegian offshore petroleum industry is being regulated by some standards and
regulations briefly described in Tables 3 and 4.

Table 3: Requirements from Norwegian regulations

Provisions Relevant to Gas Detection Norwegian Guidelines Related Standards

Safety barriers: Safety functions being regarded Sections 3 and 8 of the Facil-
as barriers against hazards and accidents ities and 4 and 5 of the Man-
agement Regulations in PSA
Guidelines
Design of safety functions: Requirements for de- Section 8 of the Facilities NS-EN ISO 13702, NOR-
sign of safety functions Regulations in PSA Guide- SOK S-001 and IEC 61508
lines, OLF 070 Guideline
Design of fire and gas system: Requirements for Section 8 of the Facilities NS-EN ISO 13702 with Ap-
design of fire and gas detection systems Regulations in PSA Guide- pendix B.6, NORSOK S-
lines, OLF 070 Guideline 001 Chapters 12 and 13
Disconnection: When it becomes necessary to Sections 8 of the Facilities
disconnect safety functions, the requirements and 26 of the Activities Reg-
shall be applied. ulations in PSA Guidelines
Performance requirements: Performance re- Section 1 and 2 of the man- IEC 61508
quirements shall be established for all safety bar- agement regulations in PSA
riers on an installation Guidelines, OLF 070 Guide-
line
Availability: The requirement for available status Section 8 of the Facilities NORSOK I- 002, Chapter 4
shall be fulfilled. Regulations in PSA Guide-
lines
Independence: The fire and gas detection system Section 32 of the Facilities
shall come in addition to systems for management Regulations in PSA Guide-
and control and other safety systems. lines
Interface: The fire and gas detection system may Section 32 of the Facilities
have an interface with other systems as long as Regulations in PSA Guide-
it cannot be adversely affected as a consequence lines
of system failures, failures or isolated incidents in
these systems.
Limiting consequences: Relevant safety func- Section 32 of the Facilities
tions shall be activated when there is a demand Regulations in PSA Guide-
on the detection system lines
Not Permanently Manned Facilities: They Section 32 of the Facilities
should also have a dedicated gas detection func- Regulations in PSA Guide-
tion for the area around and on the helicopter deck lines
Visual perception of detection: Detection of gas Section 32 of the Facilities
should be shown by means of a light signal that is Regulations in PSA Guide-
visible at a safe distance from the facility. lines
Gas detection for mobile units: For mobile fa- Section 32 of the Facilities DNV-OS-D301 Chapter 2,
cilities that are not production facilities, and that Regulations in PSA Guide- Section 4, subsection D
are registered in a national ships’ register lines

The role of gas detection as stipulated in NORSOK S-001 (subsection 12.1) shall encompass the con-
tinuous monitoring of flammable or toxic gases. The standard focused primarily on hydrocarbon (HC)
gas detection (including H2 as relevant), H2 S gas detection, CO2 gas detection and CO gas detection
wherein it sets alarm limits for each of these. For hydrocarbon gas detection (including H2 as relevant),
i.e. flammable gas detection, the alarm limits (both low and high) are fixed in relation to the types of de-
tectors in use whether point detectors or IR open path detectors. It is possible to use a single alarm limit

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
 Table 4: Requirements from Norsok S-001 Standard

Provisions NORSOK S-001 Requirements References

Role Continuous monitoring of flammable or toxic gases. subsection 12.1
Interfaces Link between gas detection system and ESD, BD system, ISC, ventilation, PA subsection 12.2
and alarms system and fire fighting systems.
Required utili- Uninterrupted Power Supply (UPS) and instrument air supply (if aspiration Subsection 12.3
ties system is applied) are required in gas detection system.
Detection design Speedy and reliable detection before gas cloud reaches critical concentra- subsection 12.4,
coverage tion/size. subsubsection
12.4.1
Leak detection All potential flammable gas leak points shall have flammable gas detection. subsection 12.4,
subsubsection
12.4.2
Herein, the smallest gas cloud with the least unacceptable consequence shall Same as above
be the basis for confirmed gas detection.
In naturally ventilated area, a smaller leak rate for warning (alarm) is enough Same as above
and is typically 0.1 kg/s.
In mechanically ventilated areas, detection of smaller leaks shall be subject to Same as above
expert judgment.
Deploying detectors shall be based on an assessment of gas leak scenarios in Same as above
relation to potential leakage source and rate, dispersion, density, equipment
arrangement, ventilation and the probability of small leak detection therein.
The basis for selection and placement of detection in each area shall be docu- Same as above
mented.
Open path detectors are preferred where the layout enables good coverage by Same as above
them.
Detection principle to apply shall be subject to considerations for environmen- Same as above
tal conditions and availability of protection for detectors.
Catalytic detectors shall not be used unless other detectors do not perform as Same as above
required.
Detection loca- Sufficient detectors shall be located by natural passageways along flow direc- subsection 12.4,
tion tion, in different levels in an area or module, in potential gas traps and in the subsubsection
air inlets of heat sources and accessible without scaffolding. 12.4.3
Detection char- The detector characteristics and calibration shall guarantee good estimation subsection 12.4,
acteristics and for gas concentration (point detectors), gas amount (open path detectors) or subsubsection
calibration leakage rate (acoustic detectors) 12.4.4
Detection ac- The detection system shall activate all actions according to the Fire and Explo- subsection 12.4,
tions and voting sion Strategy (FES). subsubsection
12.4.5
Detection levels Detectors used shall give alarms as soon as possible and within the recom- subsection 12.4,
mended alarm limits/settings. Detection, failure of further action on demand subsubsection
and system defect shall be reflected in central control room (CCR) as alarms. 12.4.6
Use alarm limits and outputs for annunciation as stipulated by standard.
Detection re- Maximum response time of detection shall be defined so as to ensure fulfill- subsection 12.4,
sponse time ment of total reaction time for each safety function. Apply recommended re- subsubsection
sponse times unless reduction is needed 12.4.7
Detection logic Logic solver compliance with the intended use and safety integrity requirement subsubsections
solver shall be demonstrated. 12.4.8 and 9.4.6
and IEC61508
Fire and gas in- The fire and gas detection system shall operate as an independent system. subsection 12.4,
dependence subsubsection
12.4.9
Survivability re- The gas detection system shall not be dependent on local instrument rooms subsection 12.5
quirements with location less safe than the central control room.
Equipment critical to effectuation of system actions shall be protected against subsection 12.5
mechanical damage and accidental loads until all actions from the detection
system have been activated.

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
for hydrocarbon gas detection, but this must be the low alarm limits. The alarm limits for area detection
systems (e.g. ultrasonic/acoustic detectors) are left to the operators to decide and adjust on the basis of
the background noise peculiar to their operating environment. However, guidelines on the use of IR gas
cloud imaging, a type of area detection, has yet to be treated by the standard. This is probably due to
its being a new technique that has yet to be applied extensively in the Norwegian industry. Furthermore,
the alarm limits for toxic gases are defined in the standard based on the effect of toxic gas in relation
to concentration or exposure time and these vary for H2 S gas detection, CO2 gas detection and CO gas
detection.

5. COMPARATIVE STUDY OF PERFORMANCE STANDARDS FOR GAS DETEC-

TION WORLDWIDE

The performance standards for gas detection do specify the performance levels to which gas detectors
should be tested and operated, and several variations of these exist across the geographical regions of
the world. The variations are probably as a result of diversity of regulatory agencies. Some of the
standards available in different countries have little differences, whereas the differences between some
are significant. However, they all have a common goal which is the prevention of accidents.

In North America, as regards flammable gas detection performance specifications, FM 6310/6320 (used
mainly in the US) is similar to C22.2 152 (used mainly in Canada) and both of them are closely re-
lated to ANSI/ISA 12.13.01-2000 [4]. As regards offshore toxic gas detection, the ANSI/ISA 92.00.01
is widely used worldwide [4] and emphasizes on repeatability, step-response and recovery as part of
requirements for toxic gas detection performance tests with the worst case accessory attached [2]. The
ANSI/ISA 12.13.04 recommends instrument measurements in LEL-m (lower explosion limit meters) or
ppm-m (parts per million meters) for flammable gas open-path detection [2]. It also recommends several
rigorous tests covering solar immunity, simulated fog/mist and water vapor, partial obscuration of optics,
long range operation with 95% obscuration of optics, vibration and temperature extremes and long term
stability, either while under stress or before and after stress [2]. The ANSI/ISA 92.00.04 also demands
measurement in ppm-m (only) of the toxic gas in the optical beam of the open-path toxic gas detector as
well as a misalignment test [2].

In Europe, the national standards are becoming harmonized with the European standards and the IEC
standards. For example, as regards both point and open-path flammable gas detection performance
specifications, the IEC 60079-29 series have been adopted by many European countries [4]. The same
also applies for IEC 45544 series which are dedicated to toxic gas detection [4]. The IEC/EN 60079-
29 series recommends that a detector for flammable gas should be used where the accumulation of a
combustible air-gas mixture can pose a hazard to life and assets. Furthermore, such a detector is required
to sound alarms, show visual warnings or initiate mitigative actions. In addition, IEC/EN 60079-29 and
IEC/EN 45544 series advise on considering the effects of variations in temperature and humidity of the
gas marked for detection.

In Norway, NORSOK standards which are developed by the Norwegian petroleum industry are widely
in use. They are a range of standards intended to serve as references or bases upon which relevant Nor-
wegian regulatory bodies can prescribe statutory requirements and evaluate their compliance. In addi-
tion, NORSOK standards serve as replacements for oil company specifications and they normally make
necessary additional provisions to recognized international standards in order to address some needs
peculiar to the Norwegian petroleum industry [17]. The NORSOK standard that treats gas detection is
NORSOK S-001 (Technical safety) which has been described to some extent earlier. NORSOK S-001
is an all-in-one standard generally covering point, open-path and area detection of both flammable and
toxic gases. This is unlike the other standards that are separated such that each covers not more than one

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
of the following aspects: flammable-gas point detection (IEC/EN 60079-29-1 and ANSI/ISA 12.13.01),
toxic gas point detection (IEC/EN 45544 series and ANSI/ISA 92.00.01), flammable gas open-path de-
tection (IEC/EN 60079-29-4 and ANSI/ISA 12.13.04), toxic gas open-path detection (IEC/EN 45544
series and ANSI/ISA 92.00.04) and area detection of flammable or toxic gas.

A table briefly juxtaposing performance specifications for toxic-gas point detection across Norway, Eu-
rope and America is shown in Table 5.

Table 5: A brief comparison of performance standards for toxic-gas point detection

Toxic gas detection NORSOK S-001 IEC/EN 45544 series ISA 92.00.01 to 92.06.01,
specifications FM 6341, NFPA 70
Gas concentrations 0, 20%, 50%, 90% of full 10 to 100 ppm H2S
scale
Temperature range -10 to 40o C 14 to 122o F (–10 to 50o C)
Relative humidity range 20% RH, 50% RH, 90% 15 to 90%
RH
Response time T90 < 2 seconds T50 < 60 seconds, T90 < T20 < 10 seconds, T50 <
2.5 minutes 30 seconds
General alarm limits Maximum is 10 x 10- 70 Db(A) at 0.3 meters
6 /20 x 10-6 (low/high from apparatus
for H2S), maximum is
5000 x 10-6 /15000 x
10-6 (low/high for CO2),
maximum is 30 x 10-6
/200 x 10-6 (low/high for
CO)
Accuracy/Linearity 0.3% (for 0.5 STGC to 10% of applied gas con-
10 STGC) to 0.5% ( centration or 3 ppm
for 0.1 STGC to 0.5%
STGC)

6. CONCLUSION AND RECOMMENDATIONS

According to recent statistics, a significant percentage (about 44%) of gas releases remain undetected in
spite of the application of the detection technologies in use [19]. The main objective of the paper has
been to present a state-of-the-art knowledge of gas detection for offshore application.

This paper has given more insights into the various aspects of applicable gas detection and will be useful
to students and practitioners in offshore petroleum related fields. The paper has reviewed literature,
standards and guidelines in relation to gas detection in the offshore oil and gas industry. It has covered
the description of the various gaseous releases, the applicable detection technologies and their pros and
cons as well as standards and guidelines being applied in the offshore industry in Norway and worldwide.
In addition, a comparative study of performance requirements across international boundaries has been
done.

Based on the aforementioned, it can be inferred that no single detection technology is a complete solution
to offshore gas detection. There is the need to link various technologies together in order to achieve
complete coverage and enhanced redundancy. In this way, detection layers of protection (barriers) will
be established and made independent. This will enhance the prevention of major accidents characterized
by fire, explosion and toxic release. Besides, the associated flammable and hazardous gases, of which
exposure is inevitable, need continuous monitoring since the processes generating them are continuous.

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
Furthermore, it has been seen that no single detection standard across the world areas can be regarded
as "‘the standard of everything about gas detection offshore"’. Hence, there is the need for continuous
improvement as regards the harmonization of standards.

References

[1] Analog Devices. ADI Gas Detector Solution Based on NDIR and PID, Project Code: APM-Gas-
Detection (EC)-2012. Technical report, Analog Devices, 2012.
[2] Shankar Baliga. ISA Performance Standards for Gas Detection. Technical report, General Moni-
tors, Lake Forest, 2012. URL http://www.isa.org/~oranc/DOC/Meetings/2012/11\_nov/
isa\_performance\_standards\_for\_gas\_detection.pdf.
[3] RW Bogue. Technology Roadmap: Optoelectronic gas sensors in the Petrochemicals, Gas and
Water industries. Technical report, OptoCem.Net, 2006. URL www.optocem.net.
[4] Bill Crosley and Simon Pate. Survey of Standards Related to Gas Detectors. Technical report,
Minneapolis, 2011. URL http://safety.det-tronics.com.
[5] Dräger. Introduction to Gas Detection Systems. Technical report, Dräger Safety AG & Co. KGaA,
2008. URL http://www.draeger.net/media/10/03/05/10030546/introduction\_gds\
_fl\_9046421\_en.pdf.
[6] Dräger. Multi-gas detection, 2013. URL http://www.draeger.com/sites/enus\_us/Pages/
Fire/X-zone-5000.aspx?navID=427.
[7] Trevor Kletz. Learning from accidents. Gulf Professional Publishing, Oxford, third edition, 2001.
[8] S.W. Legg, C. Wang, a.J. Benavides-Serrano, and C.D. Laird. Optimal gas detector placement un-
der uncertainty considering Conditional-Value-at-Risk. Journal of Loss Prevention in the Process
Industries, 26(3):410–417, May 2013. ISSN 09504230. doi: 10.1016/j.jlp.2012.06.006. URL
http://linkinghub.elsevier.com/retrieve/pii/S0950423012000952.
[9] Stuart McAleese. Safety Procedures. In Operational aspects of oil and gas well testing, chapter 4,
pages 19–43. Elsevier, Amsterdam, 1st ed edition, 2000.
[10] MSA. MSA Gas Detection Handbook. Mine Safety Appliances Company (MSA), Pittsburg, 5th
ed. edition, 2007.
[11] Edward Naranjo and Gregory A Neethling. Integrated Solutions for Fixed Gas Detection. Tech-
nical report, General Monitors, 2010. URL http://www.gassonic.com/fileadmin/user\
_upload/documents/GM-Gassonic\_Whitepaper\_Final.pdf.
[12] P W Nebiker and R E Pleisch. Photoacoustic gas detection for fire warning. Fire Safety Journal,
37:429–436, 2002.
[13] Dennis Nolan. Characteristics of Hydrocarbon Releases , Fires , and Explosions. In Handbook of
Fire and Explosion Protection Engineering Principles, chapter 5, pages 49–69. Elsevier Inc., 2011.
ISBN 9781437778571. doi: 10.1016/B978-1-4377-7857-1.00005-7.
[14] Dennis Nolan. Fire and Gas Detection and Alarm Systems. In Handbook of Fire and Explo-
sion Protection Engineering Principles, chapter 17, pages 181–201. Elsevier Inc., 2011. ISBN
9781437778571. doi: 10.1016/B978-1-4377-7857-1.00017-3.
[15] Oil & Gas UK. Atmospheric emissions, 2009. URL http://www.oilandgasuk.co.uk/
knowledgecentre/atmospheric\_emissions.cfm.

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii
 [16] R. Rubio, J. Santander, J. Fonollosa, L. Fonseca, I. Gràcia, C. Cané, M. Moreno, and S. Marco.
Exploration of the metrological performance of a gas detector based on an array of unspecific in-
frared filters. Sensors and Actuators B: Chemical, 116(1-2):183–191, July 2006. ISSN 09254005.
doi: 10.1016/j.snb.2006.03.018. URL http://linkinghub.elsevier.com/retrieve/pii/
S0925400506001821.
[17] Standards Norway. Technical Safety. Technical report, Standards Norway, Lysaker, 2008. URL
www.standard.no/petroleum.
[18] S Stueflotten, T Christenser, S Iversen, J Hellvik, K Almå s, T Wien, and A Graav. An Infrared
Fibre Optic Gas Detection System. In Proc. SPIE. 0514, 2nd Intl Conf on Optical Fiber Sensors:
OFS’84 87, pages 87–90. SPIE, 1984. doi: 10.1117/12.945064.
[19] AM Thyer. Offshore Ignition Probability Arguments, Report Number HSL/2005/50. Technical
report, Health and Safety Laboratory, Buxton, 2005.
[20] J.E. Vinnem, R. Bye, B.A. Gran, T. Kongsvik, O.M. Nyheim, E.H. Okstad, J. Seljelid, and J. Vatn.
Risk modelling of maintenance work on major process equipment on offshore petroleum installa-
tions. Journal of Loss Prevention in the Process Industries, 25(2):274–292, March 2012. ISSN
09504230. doi: 10.1016/j.jlp.2011.11.001.

Probabilistic Safety Assessment and Management PSAM 12, June 2014, Honolulu, Hawaii

View publication stats