(509869149) Leak Detection
Content
Advance Surveillance System for Natural Gas Pipeline Leakage Detection using Fibre Optics Distributed emperature Sensing Sat!is! "yengar Certification Engineers India Limited, Mumbai, India
Abstract
Natural Gas Pipelines have been a vital component of the energy supply chain in India; have to be laid in harsh surroundings; crossing mountain ranges characteried characterie d by unstable grounds; !here seasonal soil te"ture changes increase the probability of haards haards and uncertainties# $herefore, $herefore, Natural Gas pipeline monitoring systems for lea%age, ground movement, and intrusion detection are part of ne! pipeline pro&ects# Lea%age detection using distributed fibre'optic sensors can be a comprehensive solution for continuous, in'line, real' time monitoring of various pipelines#
$he monitoring of temperature profiles over long distance by means of optical fibres represents a highly efficient !ay to perform lea%age detection along pipelines# (ifferent techni)ues have been developed ta%ing advantages of the fibre geometry and of optical time domain d omain analysis for the localiation of the information# *aman'based systems systems have been envisaged for one of the very first pro&ects of India !here lea%age detection using (istributed $emper $emperature ature +ensing has been envisaged# $he paper presents and discusses the possibility to actively and automatica automatically lly monitor monitor lea%ages lea%ages using using distribut distributed ed fibre optics optics sensing sensing techni)ues# $he second part of the paper focuses on the monitoring of lea%age and third party intrusion detection of petroleum product pipelines# $he %ey features and performances of the technology are revie!ed in this paper#
Keywords: pipeline leakage detection, intrusion detection and temperature monitoring, Raman Scattering, fibre optics sensor, database management 1
"ntroduction In india in - ./01'02 there are t!o ma&or incidents incidents that have ta%en place in India in month of 3uly and +eptmber ./01 in the state of 4ndhra Pradesh and 4ss 4ssam# am# In 0 st incident .. people !ere %illed and . nd incident !ere 5 people !ere %illed#
6ith increasing public consciousness and concern for the environment, recent pipeline lea% incidents incidents have proved that the cost to a company company can be far more than the do!ntime and clean up e"penses# 4s m more ore stringent statutory statutory regulations regulations are getting introduced, cost effective and reliable lea% detection systems are in demand#
$he paper paper present presents s and discusses discusses the possibil possibility ity to actively actively and automati automatical cally ly moni monitor tor lea%ages using distributed fibre optics sensing techni)ues# $he second part of the paper focuses on the monitoring of lea%age and third party intrusion detection of ./ petroleum petroleum product pipelines pipelines !ith lengths l engths varying from 7'0/ %ms appro"# $his is one of the very first pro&ects of India !here lea%age detection using (istributed $emperature +ensing has been envisaged#
(istributed temperature sensing systems 8($+9 are o pto el e ec le ctron c ic devices !hich measure ical al fib ibre res s functioning as linear se ns or s# s# $emperatures are temperatures by means of optic recorded along the optical sensor cable, thus not at points, but as a continuous continuous profile# 4 high accuracy of temperature determination is achieved over great distances# $ypically the ($+ systems can locate the temperature temperature to a spatial resolution of 0 m !ith ac accuracy curacy to !ithin :0C at a resolution of /#/0C <0, 2=#
Physical measurement dimensions, dimensions, such as temperature or pressure, can affect glass fibres and locally change the characteristics of light transmission in the fibre# 4s a result of the damping of the light in the )uart glass fibres through scattering, the location of an e"ternal physical effect can be determined so that the optical fibre can be employed as a linear sensor# >ptical fibres are made from doped )uart glass# ?uart glass is a form of silicon dio"ide 8+i>.9 !ith amorphous solid structure# $hermal effects induce lattice oscillations !ithin the solid# 6hen 6 hen light falls onto these thermally e"cited molecular oscillations, an interaction occurs bet!een the light particles 8photons9 and the electrons of the molecule# Light scattering, also %no!n as *aman scattering, occurs in the optical fibre <1, 2=#
2
$he *aman *aman scattere scattered d light light is caused caused by thermal thermally ly influ influence enced d molecula molecularr vibration vibrations# s# Conse)uently the bac%scattered light carries the information on the local temperature !here the scattering occurred# In fact the *aman bac%scattered bac%scattered light has t!o fre)uency shifted components@ the +to%es and the 4nti'+to%es components <0=# $he amplitude of the 4nti'+to%es component component is strongly temperature temperature dependent dependent !hereas the amplitude of the +to%es component is not# $herefore *aman sensing techni)ue re)uires some filtering to isolate the relevant r elevant fre)uency fre)uency components and consists in the recording of the rratio atio bet!een 4nti'+to%es amplitude by the +to%es amplitude, !hich contains the temperature information# igure 0 sho!s the spectrum of the scattered light in optical fibres assuming that a single !avelength Ao is launched in the fibre# Brillouin'based sensing techni)ues rely on the measurement measurement of a fre)uency as opposed opposed to *aman' bas based ed techni)ues techni)ues !hich are intensity based <0=# Pro&ect defined in this paper, adopts *aman based scattering for sensing, as ma"imum pipeline length is limited to 0/ %ms#
Figure-1 Schematic representation representation of the scattered light spectrum from a single wavelength signal propagating in optical fibres. An increase of the fibre temperature has an effect on the both Raman and Brillouin components
$he temperature measuring system consists of a controller 8laser source, optical module, mi"e "er, r, receiv receiver er and and micro' micro'pro proces cessor sor unit9 unit9 and and a )uart )uart gla glass ss fib fibre re as lin line'sh e'shap aped ed temperature sensor 8figure .9# $he fibre optic cable is passive in nature and has no individual sensing points and therefore can be manufactured based on standard telecom fibres# Because the system designerD integrator does not have to !orry about the precise location of each sensing point, the cost for designing and installing a sensing system based on distributed fibre optic sensors is reduced from that of traditional sensors <1=# 4dditionally, because the sensing cable has no moving parts and design life of more 3
than 5/ years, the maintenance and operation costs are also e"pected to be considerably less than for conventional sensors# 4dvantages of having fibre optic sensing technology includes large number of monitored points over a single optical fibre sensor,, immunity to electromagnetic sensor electromagnetic interference, interference, vibration, insensitiveness to humidity and corrosion, no active electronic circuits along the cable, long'term reliability and is safe for use in haardous ones 8the laser po!er falls belo! the levels that can cause ignition9, thus ma%ing these sensors ideal for use in industrial sensing applications <0, 1, 7=#
Figure 2 – Schematic arrangement for light traveling through fibre
Pro#ect Definition Pro&ect targets to detect lea%ages along the !hole length of the pipelines to increase %no!ledge, to plan maintenance interventions interventions and to ensure safety# $he monitoring parameters are average temperature distribution and lea%age detection of various petroleum product pipelines pipelines !ith lengths varying from 7'0/ %ms appro"# feeding &etty# Lea% detection for fluids li%e Crude oil, Naphtha, (P, M+, +(, Para"ylene, Propylene, +ervice !ater, !ater , Nitrogen, LPG, 4$, +> as depicted in figure 5, for pipelines pipeli nes ranging from 1F to 5F has been envisaged <=#
4
Figure-3 ipelines schematic
Each section consists of one temperature sensing cable !ith four single mode fibres to be buried above or belo! the pipe# Each section can be connected through optica opticall connectors or spliced together# $he pro&ect concentrates on the Continuous Monitoring#
ere the monitoring system is based on *aman scattering technolog technology, y, is selected for distributed temperature monitoring# 4s detailed above, the distributed fibre optic sensors shall detect temperature changes !ith resolutions up to !.!"# $ # +patial resolution depends depend s on sensor cable length, and is typically one meter for the present ma"imum lengths of up to 0/ %m <=#
5
System Description $ %onitoring Strategy $he system consists of reading unit, sensing cable and accessories 8connection bo"es, e"tension cables, splice protectors etc#9# $he optical fibres, !hich are integrated into robust cables, are the temperature sensitive elements and allo! the measurement of temperature profiles at arbitrary times, )uasi'continuously !ith a high spatial resolution along the cable#
Li)uid lea% detection monitoring !ill be performed indirectly belo! the pipe 8temperature cable at H >cloc% position9 by the temperature increase in the ground#
Gas lea% detection monitoring !ill be performed indirectly on the top part of the pipe 8temperature cable at 0. >cloc% position9 by the temperature decrease in the ground induced by the decompression of the lea%ing gas caused by the 3oule'$hompson effect#
Intrusion detection !ill be performed indirectly on the top part of the pipe 8temperature cable at 0. >cloc% position9 by the temperature change in case of removal of covering material#
igure 1 sho!s the typical trench layout for laying of optical fibre cables for different product lines#
Figure-% &rench 'etails $ross Section
6
ollo!ing parameters are to be monitored using this methodology@
4verage temperature along the sensor !ith !ith spatial resolution resolution of 0'. 0'. m
4verage temperature threshold detection along along the sensor sensor
Measurement of temperature variation along the sensor
Lea%age detection
$hird party intrusion detection
$he present monitoring strategy contains a certain sensor redundancy, redundancy, !hich is necessary for cases !here the sensors are damaged during installation or later# ence, even if one sensor in one cable gets damaged after a certain time, the global performance of the system !ould not be decreased# $o install the sensors at proposed location it is necessary to be sure that no physical or constructive obstacle is presented#
$he origin of the temperature disturbance disturbance around the pipeline depends on the type of pipeline and its surroundings# $he most typical effects are the follo!ing@
$he released li)uid is !armer than the surrounding soil 8typical for buried oil and li)uid pipelines9
$he released hydrocarbon li)uid changes the thermal properties of the soil, in particular thermal capacity, and influences the natural dayD night temperature cycles
Gas lea%age is detected by the temperature decrease in the ground induced by the decompression of the lea%ing gas caused by the 3oule'$hompson effect
$he above effects influence the ideal cable placement around the pipeline#
Ground temperature@ lea%age of oilD !ater is detected because of punctual temperature temperatu re increase
In the case of a buried oil pipeline the best location for the sensing cable is belo! the pipe, but not in direct contact# 4t that position there is a ma"imum probability of collecting the released li)uid, independently from the lea%age location# $he (istributed $emperature +ensing cable has therefore to be installed, appro"imately bet!een /#. m
7
and /#2 m, belo! the pipeline along its !hole length# igure' 2 demonstrates the typical pea% received at the point of li)uid lea% detection#
Figure-" &(pical pea) during li*uid lea) detection
Lea%age of gas detected because of strong punctual temperature decrease
In the case of gas lea% detection in buried pipelines the best location for the sensing cable is above the pipe# 4t that position there is a ma"imum probability of collecting the released gas, independently from the lea%age location# igure'H demonstrates the typical pea% received at the point of gas lea% detection# #
Figure-+ &(pical pea) during gas lea) detection
8
Intrusion detection detected by temperature change in case of removal of
covering material 8J$ K $0'$.9 as depicted in figure 7 !here J$ is the difference of $0 and $. !hich is detected by fibre optic cable#
Figure-, ntrusion 'etection
$he system can detect the removal of earth from the optical fibre cable# $his results in an immediate change in the recorded temperature that can be used to generate an alert# $he position of the event can clearly be identified in all situations#
Intrusion detection through temperature anomalies analysis
Change of cable temperature due to digging and cable e"posure
Change of pipeline temperature due to e"posure to air
•
•
•
System Features Ma&or system re)uirementsD features for the pro&ect include@ *eading unit !ith data ac)uisition soft!are to sho! the results locally and remotely and in form of !arnings and pre'!arnings depending on the measurements# (istributed (ata Management and 4nalysis +oft!are' an integral and fully compatible part of distributed monitoring system for data storing, processing, representation and analysis, as !ell as for the control of single or multiple reading units# $he main functions of the soft!are are aimed to measure sensors automatically# $he operator shall vie! in real time the sensors measurement history in graphical form# +oft!are shall provide the platform to monitor various trends, graphs for the entire length of pipeline as depicted in figure'H and 7# $he soft!are shall trigger alerts 8+M+, mail and phone call9 and sho! !arnings on the display# $he soft!are shall combine measurements from different sensing cables to obtain comple" results# $he soft!are stores all information related to a sensor in a single
9
data'base structure# 4ll data to be e"ported to third party soft!are is depicted in figure including M+ E"cel and M+ 4ccess# Multiple users to access the soft!are simultaneously from different PC 8locally or remotely over a modem or L4N9#
Figure- 'ata &ransfer in '&S
4s continuous continuous monitoring is vital to this pro&ect, alarms to be classified as non' threat, possible threatD lea%, and a threat D lea%# Each event classification to be colour coded 8i#e# green no threat D lea%, yello! possible threat D lea%, red 8flashing9 threat D lea%9 for easy identification# Intelligent 4larms ' $he soft!are shall also include assignment of ones to each pipeline varying in length as depicted in figure # It shall be possible to change the sensitivity or isolate alarm and events based on the pipeline one# Each one can be individually tuned to the local environmental conditions and have parameters set to distinguish the differences in the identification of possible noises#
10
Figure-/ 'ifferent Alarm $riteria
&onclusion *ecogniing the importance of lea% detection in the prevention of oil spills and the need for a more thorough understanding understanding of the use and effectiveness effectiveness of lea% detection technologies technolo gies has led ma&or oil companies to adopt the best possible technologies technologies available# >ften it is difficult for a pipeline company to discern, !hat is the best solution for their particular pipeline and philosophy of operation# (istributed $emperature +ensing is one of the prominentD emerging technologies !hich offer several advantages and posses clear advantages over other e"isting conventional sensors# 6ith this pro&ect a ne! initiative has been ta%en#
'eferences( 0# Marc Ni%les, Bernhard Bernhard ogel, ogel, abien abien Briffod, +tephan +tephan Gross!ig, lorian lorian +auser, +auser, +teffen Luebbec%e, 4ndrO Bals, $homas Pfeiffer ' Proceedings of the 00th +PIE 4nnual International 4nnual International +ymposium +ymposium on +mart +mart +tructures and Materials, Materials, March 01'0, 01'0, .//1, +an (iego, California, +4, 2. 'r 0un 1hang 'esigning a $ost ffective and Reliable ipeline 4ea) 'etection S(stem 5# E# $apanes, ibre optic sensing solutions for real time pipeline integrity monitoring 1# 'r. Stuart 4. Scott 'r. 5aria A. Barrufet ' 6orld!ide 4ssessment 4ssessment of Industry Industry Lea% (etection Capabilities for +ingle Q Multiphase Pipelines, .//5 2# 'an 'aniele iele naudi naudi and Bran)o 6lisic 6lisic Fibre Fibre 7ptic 7ptic Sensing Sensing for nnovati nnovative ve 7il 8 6asroduction and &ransport S(stems
11
+. (aniele Inaudi, Bran%o Glisic, 'istributed Fiber optic Strain and &emperature Sensing for Structural 9ealth 5onitoring ,. (a!n # Gifford, Brian 3# +oller, Matthe! +# 6olfe, Mar% E# roggatt' (istributed iber'>ptic $emperature +ensing using *ayleigh Bac%scatter . Engineering (esign (ocument, +outh 3etty Pro&ect, EIL, Ne! (elhi
12
