0% found this document useful (0 votes)
242 views

Webinar - Specification of Passive Fire Protection Using Phast, Safeti and CFD

DNV Passive fire protection
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
242 views

Webinar - Specification of Passive Fire Protection Using Phast, Safeti and CFD

DNV Passive fire protection
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 31

Webinar: Specification of passive fire protection

using Phast/Safeti and CFD


2 April 2019

1 DNV GL © SAFER, SMARTER, GREENER


Your presenters

James Pickles Rune Natten Kleiveland


Product Owner Phast and Safeti CFD Product Manager / Principal
Specialist

2 DNV GL ©
Agenda

▪ Background to Passive Fire Protection in the oil & gas industry

▪ How we can use Phast/Safeti to identify high escalation risk areas – a screening analysis

▪ How we can use CFD for detailed analysis / optimization

▪ Questions

3 DNV GL ©
Background in the oil & gas industry

▪ What is escalation?
– An event that becomes worse due to release of additional inventory
▪ Passive fire protection is a barrier to escalation

▪ Too little: high escalation risk from jet fires or pool fires

▪ Too much: high cost of installation and maintenance

▪ Optimal: define acceptance criteria to focus efforts on protecting areas


most exposed to escalation risk

▪ Poll question

4 DNV GL ©
Hypothetical problem

▪ Passive fire protection across a 40-year old gas terminal is Muster


point
beginning to degrade.

▪ The operator does not understand the original basis for PFP at the
terminal – there doesn’t seem to be a philosophy.

▪ Operator wants a clear philosophy on PFP and to know which PFP


they should maintain/replace and which they could remove.

Control
room

5 DNV GL ©
The solution

▪ Utilise the site’s QRA model (in Safeti) to identify the areas with the highest jet fire and pool fire
escalation risk

▪ Assess the consequences of escalation

▪ Combine this information to provide guidance on the areas where PFP would provide the most
benefit

6 DNV GL ©
How do we do it?

Muster
▪ First, define the rulesets: point

▪ Risk-based approach

▪ Apply/maintain PFP where escalation frequency > 10-4 per


year AND escalation would result in fatalities at either the
control room or the muster points

▪ Where the escalation frequency is between 10-4 and 10-6 per


year AND escalation would result in fatalities at either the Control
room
control room or the muster points, apply/maintain PFP
where it is reasonably practicable.

7 DNV GL ©
How do we do it?

▪ Then, apply the rulesets:


▪ Risk effect levels in the QRA model.
– Provides frequency contours for a specified thermal radiation

For example: frequency contours for 180 kW/m2

8 DNV GL ©
How do we do it?

▪ Escalation scenario assumed to involve impingement of a flame on pipework or equipment (e.g. a


vessel) causing rupture of the pipework or release of the contents of the equipment
instantaneously.
▪ Identify the escalation scenarios that could cause fatalities at either the control room or the
muster point.

Escalation scenario Distance to Fatalities in Escalation risk Escalation risk


fatality control room > 10-4 per > 10-6 per
criterion or at muster year year
point
Rupture of slug catcher 8” pipework 90 m No Yes Yes

Separator BLEVE 150 m Yes Yes Yes

Rupture of separator 12” pipework 120 m Yes No Yes

Suction scrubber BLEVE 130 m Yes No No

Rupture of compressor 10” pipework 170 m Yes Yes Yes

9 DNV GL ©
What does this give us?

▪ Prioritization / screening of escalation scenarios / equipment that would benefit from PFP

▪ Operator wants a clear philosophy on PFP and to know which PFP they should maintain/replace
and which they could remove.

▪ This information can be used as a basis for CFD analysis

10 DNV GL ©
Further possibilities

▪ We have not discussed the impact of time


– The risk plots are based on fires at t = 0
(conservative)
– Fire size likely to decrease with time.

– Define the duration of a fire required to cause


escalation (e.g. 10 minutes)

– Setup time-varying leak models and specify the “time at which release rate is calculated”
– The risk effect contours will be based on the fire sizes at the specified time.

11 DNV GL ©
Further possibilities

▪ We have not discussed the impact of isolation and blowdown


– Setup time-varying leak models and specify isolation
and blowdown capability
– The risk effect contours will then factor in this
capability

Isolation Blowdown

12 DNV GL ©
Further possibilities

Without blowdown @ escalation time With blowdown @ escalation time

13 DNV GL ©
Further possibilities

▪ This provides a quantified benefit of the isolation and blowdown system, leading to a reduced
need for PFP.

▪ Similarly, it identifies the equipment that would still benefit from PFP even in the presence of a
fast performing isolation and blowdown system.

14 DNV GL ©
Summary

▪ Use Safeti to show frequency contour plots (“risk effect levels”) for a given thermal radiation level
– Provides an indication of the escalation risk

▪ Identify high risk areas

▪ Determine consequences of escalation

▪ Based on defined criteria, make recommendations for PFP

▪ Further detailed analysis using CFD

▪ Poll question

15 DNV GL ©
Optimization of passive fire protection (PFP)

Case study

16 DNV GL ©
Why apply CFD? (Computational Fluid Dynamics)
▪ CFD is especially interesting to solve near field problems;

▪ The congestion/confinement of an installation plays important role in fire results;

▪ Obstruction can prevent radiation from reaching other areas and/or increase the radiation
locally, leading to more accurate results.

▪ KFX™:
– CFD simulator for fires and gas dispersion
in complex geometries.

17 DNV GL ©
Why spend money optimising PFP design?
▪ More PFP does not always increase safety level.
– PFP makes it difficult to inspect piping and
equipment.
– Corrosion and wear can be difficult detect.
– Significant weight increase.

– (Can increase heat loads to unprotected objects) Sufficient amount of


Typical amount of
PFP used on North
Sea platforms"
PFP to achieve the
I ncreasing! required safety level"
Safety"

▪ A detailed analysis may show that additional fire Acceptable!

protection is required in certain areas.


Safety Level"

“Waste of M oney”

▪ Up to 70% reduction in amount of PFP if designed Amount of passive fire protection, PFP"

with an advanced methodology compared to


prescriptive methodology. Advanced Fire Design

18 DNV GL ©
Case study - the problem

▪ New built offshore production platform.


▪ Design of Passive Fire Protection (PFP) layout established in early design stage.
– Based on a simplified analysis
– All deck beams to be fully protected
– All critical pipe supports to be protected

▪ Process layout and piping changed during engineering and building phase.
▪ Delays during building phase
▪ At time of delivery:
– No PFP on deck beams
– No PFP on pipe supports
– PFP on some of the vessel supports

19 DNV GL ©
Objective

▪ Investigate whether un-protected support structures for piping and process equipment can
withstand heat loads from dimensioning fires.

▪ Which structures will need additional fire protection?

▪ Detailed study focusing on specific critical areas.

▪ Could typically be a continuation of a Safeti QRA based PFP study as presented earlier.

20 DNV GL ©
Conventional methodology

▪ Use dimensioning accidental loads from regulations:


– NORSOK S-001:

▪ First assumption:
– All parts of the module may experience the same high heat load, at the same time.
– Same intensity during the entire fire scenario.
– Conservative approach

21 DNV GL ©
Methodology – Part 1 - Fire simulations with KFX™
▪ Selection of fire scenarios.
– (Representative selection, impossible to
simulate everything)
– From a risk analysis (QRA)
– Probabilistic method.
– Scenario based.
– Selection:
– Leak position.
– Leak rate.
– Release direction.
– Composition.
– Wind conditions.

DNV GL ©
Methodology – Part 1 - Fire simulations with KFX™

▪ Captures the time development. Pool fire,


▪ Captures the interaction between fire and scenario 3
structure.
– Impingement.
– Buoyancy. After 3
minutes
▪ Detailed representation of heat loads to
structures and process equipment.
– Shielding.
– Radiation and convective heat loads.
– Region with high heat loads
– Part of object which experience high heat After 30
loads. minutes

DNV GL ©
KFX™ Methodology – Part 2

▪ Heat loads from KFX™ as input to a structural response simulation


– KFX™: High resolution heat load data for realistic fire scenarios
– USFOS: Calculation of resulting temperature increase (Fahts) and the structural response
– Buckling, sagging, collapse, etc.
KFX™ USFOS

24 DNV GL ©
Results, KO-drum supports, Mechanical response Fahts/Usfos

Temperature response Mechanical response

DNV GL ©
Pipe supports, Mechanical response Fahts/Usfos

Temperature response Mechanical response

DNV GL ©
Design criteria – rule set

▪ Yielding and failure of the pipe support itself is not a design criterion for the PFP study
– Avoid escalation

▪ The fire simulations have demonstrated that the extent of the credible fires is limited.
– Limited number of support structures that will fail for one scenario.

▪ The process piping in this module can handle far longer free spans
– Generally not necessary to protect the supports.
– (Assess pipe connections, welded vs flanged.)

27 DNV GL ©
Results – Ruleset for pipe supports

▪ Overall result: Pipe supports do not need PFP

▪ For the following special cases the supports


should be protected with PFP:
– Always for safety critical components (ESV
etc. )
– Close to pipe intersections.
– Where heavy objects are attached to the
pipe.

Very limited amount of additional PFP required


Significantly less PFP than original design

28 DNV GL ©
Summary – PFP case study

▪ ”State of the art” simulation tools for fire scenario and structural response.
– KFX™ and FAHTS/USFOS
▪ Detailed representation of the fire, thermal fluxes, affected area, and the structural
consequences.
▪ Make an overall strategy for use of PFP based on realistic fire scenarios and the actual capacity of
the structure.
▪ Apply PFP where it is required
– To avoid escalation.
▪ PFP solution optimized specifically for this process module
▪ Significant reductions in amount of PFP compared to traditional methods for PFP design.

▪ Complementary to a study with Safeti

29 DNV GL ©
PFP design – Safeti and KFX methodology

▪ Combination of Safeti and KFX


– Efficient methodology for design and optimization of passive fire protection

1. Safeti study
– QRA
– Identify dimensioning fire scenarios.
– General PFP design, based on Safeti risk effect levels.
– Identify critical areas, safety and cost.

2. KFX analysis for optimization of PFP in the critical areas


– Together with Usfos

30 DNV GL ©
www.dnvgl.com

The trademarks DNV GL®, DNV®, the Horizon Graphic and Det Norske Veritas®
SAFER, SMARTER, GREENER are the properties of companies in the Det Norske Veritas group. All rights reserved.

31 DNV GL ©

You might also like