Marine_Structural_Design Episode 15 potx

538 Part V Risk Assessment Install Fire and Blast Barriers Escalation caused by explosions can be limited by fire and blast baxriers between modules and areas. However, the barriers themselves may cause problems for keeping ventilation and introduce more blocks. The constructionhepair of such barriers may involve extensive hot work. This measure is therefore more effective in the early design stage. Active Deluge on Gas Leakage Leakage may be deluged out without causing any explosions or fires. Deluge may be particularly effective in preventing so-called runaway flame accelerations. It may also lead to a reduction of the peak overpressure. The most critical aspect in the use of deluge is that it must be triggered prior to ignition, e.g. on detection of a gas leak. Modeling of ignition has shown that the most likely interval between release and ignition is two to three minutes. Thus deluge activation has to be within the first half minute in order to be effective. Improve Resistance of Equipment and Structures The last possibility of reducing an explosion consequence is to improve the resistance of equipment and structures to blast loads. However, it is not cost-effective to design structures for the worst explosion case. Therefore this approach may be quite expensive. 30.4 FireRisk In the offshore risk assessment, usually two types of fire risk are considered: the topside fire and the fire on sea. The following sections mainly deal with the topside fire. Further, the smoke effect analysis and the structural response under the fire are normally integrated into the fire risk assessment. The distinction between what is classified ‘fire’ and what is called ‘explosion’ is relatively subjective. The total loss of the fixed platform ‘Piper Alpha’ was initiated by a small explosion, but the damage was primarily due to fire. 30.4.1 Fire Frequency Fire frequency analysis is very similar to explosion frequency analysis. The overall fire frequency can be expressed as: (30.4) PFp = Phk * PCc ‘Ignition where, PFp = Frequency of fire Pkak PCc P,gn,,jo,, =Probability of ignition = Probability of gas leakage = Probability of gas concentration The flammable gaslair concentration range determines whether it is explosion or fire for a given ignition. Further, fire scenarios are mainly caused from the following sources: blowout, riser failure, pipeline failure, process equipment failure, and dropped object. The uncontrolled hydrocarbon flow (blowout or riser failure) is considered as the main fire risk contributor to the structures. Further, dropped objects may contribute to fire only when they lead to the Chapter 30 Risk Assessment Applied fo offshore Structures 539 rupture of hydrocarbon containing equipment. Under certain conditions, structural failure or collision impact may also lead to fires. Their final consequences are largely dependent on the escalating sequences. 30.4.2 Fire Load and Consequence Assessment A brief overview of some important aspects in the fire consequence analysis is made below. Fire Types and Characteristics Despite the fact that a fire originates from combustion reactions, the process of a fire may largely depend on the factors that are not directly involved in combustion. Fires are therefore usually separated into the following types: Jet fires Fires in running liquids Fire balls Gas fires (premixed, diffuse) Other types of fire may occur in electrical equipment or in the accommodations or on sea. These ‘non-hydrocarbon’ fires are not included here. Burgan and Hamdan (2002) gave a list of research publications on fire and explosion load characteristics, structural response analysis and performance requirements. The fire load may be converted into thermal loads (time-temperature curves) acting on the structural members. Some of the time-temperature curves are available in the literature in a form suitable for use in design. The temperature-time history for a given structural member is affected by the applied heat load, the shape of the member (for heat transfer) and the use of any passive fire protection material. Table 30.2 summarizes the main characteristics that need to be determined for these fire types. Ventilation controlled fires in enclosed units (closed or partly closed) Fuel controlled fire in enclosures Pool fires in open areas or in modulus Table 30.2 Fire Load Characteristics Duration of leak Fire Response Analysis Procedures The assessment of fire response of structures has the following calculations: 540 Pari V Risk Assessment fireloads, structural time-temperature distribution, Each of these calculations may be conducted using simplified methods or nonlinear finite element simulations. Simplified calculations may be performed in the form of hand calculations or computer spreadsheets. The weakness of the simplified calculations is its inability to account for redistribution of structural internal forces during the fire. However, the simplified calculations are normally more conservative and may be calibrated against experimental results. Smoke Effect Analysis Smoke does not affect structural elements, but it is one of the major hazards to personnel in fires, especially in oil fires. The smoke effects are e.g.: Reduced visibility, Knowledge of smoke production, smoke flow, and impact of smoke on people and facilities is available from literature, laboratory tests, and experience of real fires such as the fire on Piper Alpha platform. By proper CFD codes, the smoke effects analysis in a fire scenario can be performed, and the results can be compared to the threshold values in above three areas. Structural Response to Fire Simplified methods for structural response to fire have been derived based on results fiom fire tests and fire engineering codified methods. The sophisticated computer models are based on finite element methods, that calculate the temperature increase in a structural member based on a given temperature exposure curve and the thermal properties of the materials which are also temperature dependent. The consequences of fire include i.e., releases of hydrocarbons (combustion, radiation and convection), structural response to temperature distribution. Pain and injury to the personnel due to temperature of the smoke, Incapacitation or death due to toxic or irritating components in the smoke. ‘Minor damage’ and ‘Significant damage’ do not reflect much damage to the main and secondary structures (support structure, main deck structure, and module structure), but rather to tertiary structures and to their equipment. The higher consequences, Le. ‘Severe damage’ and ‘Total loss’, will on the other hand, involve considerable damage to the main and secondary structures. The performance requirements are applied for the protection of the primary structure and safety critical structures and systems. They are defined as strength (for structural failure) and deformation limits (to ensure that support to safety critical structures and performance of blasvfire wall are not compromised). 30.4.3 Fire Risk Reduction Fire risk reduction measures may be considered in the following four aspects, see Vinnem (1 999) for more details: Chapter 30 Risk Assessment Applied to offshore Structures 54 1 Leak prevention Imition prevention Adopt welded connections Hot work procedures Flange types with reduced leak Explosion-protected equipment Maintenance of electrical equipment probability Leak detection Gas detection Escalation mevention Installation layout Segregation of areas Fire detection Emergency system Active fire protection, e.g. deluge water system, C02 Blowdown system system, etc Passive fire protection, e.g. H-60, H-30 segregation, etc. 30.4.4 Guidance on Fire and Explosion Design A probabilistic approach has been proposed in the new NORSOK guidance documents (Pappas, 2001) and in a new engineering handbook published by Corrocean (Czujko, 2001). Walker et a1 (2002) presented a guidance document based on the risk matrix approach described in API RP 2A (21st edition). The API risk classification method has been applied to fire and explosion engineering. Methods are proposed to enable the derivation of a dimensioning explosion overpressure that may be applied to a static or dynamic analysis to assess the structure against the ductility level explosion. Two levels of explosion loading are suggested for explosion assessment by analogy with earthquake assessment. For the “Ductility Level” explosion, a performance standard such as the one below is typical: “In the case of an explosion event at least one escape route must be available after the event for all survivors. For a manned platform a temporary refuge of safe mustering area must be available to protect those not in the immediate vicinity of n explosion and to survive the event without injury?‘. For the “Design Level” explosion, it is required that the primary structure remains elastic, with the essential safety systems remaining fimctional. The explosion overpressure is the cumulative overpressure distribution for the installation, showing the probability that a given overpressure will not be exceeded. The explosion overpressure may then be expressed as a function of the return period (years). 30.5 Dropped Objects The hazards of dropped objects are mainly caused from falling crane loads. Also, various cases of crane boom fall or entire crane fall have been documented. The risk picture of the crane accidents in the North Sea shows that several fatalities have occurred when an entire crane was toppled overboard. The equipment has been damaged due to falling objects. The subsea wellheads have been damaged as a result of BOPS (Blow out Preventers) falling during exploration drilling. 30.5.1 Frequency of Dropped Object Impact The frequency of dropped object impact is defined as follows (Vinnem, 1999): 542 Load Categories Heavy or multiple drill collars Part V Risk Assessment (30.5) Load Distributions (“A) Simultaneous Normal 22.2 0 Drilling and Production Production where, Pmr Ni ‘Di PHg PFv = Occurrence probability of dropped object impact = Annual number of lifts per load category i = Probability of load dropped from crane for load category i = Probability of equipment j being hit by falling load in category i, given that the load is dropped = Probability of failure of equipment j given impact by load in category i Other Heavy (> 8 tons) Medium Heavy (2-8 tons) Light (c 2 tons) Annual Lift Number and Load Distribution Table 30.3 presents two representative load distributions for simultaneous drilling and production and for normal production. Typical numbers of crane operations per crane during one year on an installation are also shown. 0.3 0.7 27.1 33.6 50.5 65.7 Number of Lifts per year 20884 8768 Probability of Dropped Load The probability of dropped loads during operations depends on the characteristics of the load and environmental conditions. Typically, only one average frequency may be estimated, Le. an average drop fiequency per lift or per crane year. A typical frequency of dropped loads is in the order of 10E-5 to 10E-4 loads dropped per crane per year. For critical lifting operations, particular emphasis is placed on adhering to strict procedures. The dropped load frequency for this so called ‘procedure lift’ may be typically 30%-70% lower than the value for a ‘normal’ crane operation. Probability of Hitting Objects A dropped crane load may hit three types of objects. Each of them with the worst consequences is presented below. The probability of hitting is usually based on geometrical considerations reflecting the areas over which the lifting is performed. Lifting over the process area is usually prohibited by operational procedures unless special restrictions are implemented. If a load is dropped under Chapter 30 Risk Assessment Applied to off ore Structures 543 such circumstance, it may be a critical event. The probability of topside equipment being hit may be expressed as follows: (30.6) where, A,# Afof-j f,, = Area of equipment j over which loads in category i may occasionally be lifted = Total area of hydrocarbon equipment over which load category i may be lifted = Ratio of critical area to total area over which lifting is performed The probability of hitting structural components or subsea equipment can be determined in similar equations based on areas over which the lifting is performed. 30.5.2 Drop Object Impact Load Assessment In principal, two cases need to be considered regarding the falling objects from the crane: Loads that are dropped on the equipment, structures, deck, or other locations which are above the sea surface. Loads that are dropped into the sea and possibly hit structures in the water or subsea equipment on the sea bottom. The first case has only one phase, i.e. the fall through air. The second case has three phases, falling through the air, impact with the sea surface, and the fall through the water. Idealized calculations to determine the impact velocities in these three phases are briefly presented below. The drift caused by the currents may also be taken into account when calculating the most probable landing point on the seabed. Fall through the Air A falling object will accelerate towards the sea surface in accordance with the force of gravity. The impact velocity can be determined by: v, =J27qk (30.7) where, h g = Gravity acceleration = Height from which the drop occurs Impact with the Water A falling object may hit the sea surface and proceed through the water with the velocity Vz , as determined by Eq. (30.8). The integral represents the loss of momentum during the impact with the water surface. (30.8) where, M = Object mass 544 Topside equipment Structural components above or in the water Subsea equipment Part V Risk Assessment May cause loss of integrity of hydrocarbon containing equipment and possibly lead to a process fire. May cause structural failure or loss of stability or buoyancy. May cause loss of containment of production (hydrocarbon containing) equipment, possibly lead to a significant oil spill. P(t) = Impact force Fall through the Water After the impact, the object will accelerate from V, towards its terminal velocity V, in the water. (30.9) where, W 0 = Buoyancy force P A = Cross section area Cd = Gravity force (in air) = Density of sea water = Shape coefficient of the object depending on the Reynolds number It is also known that an object will tend to oscillate sideways during the fall through water. These oscillating movements are determined by the impact angle with the water surface and the external shape of the object. Bar-shaped objects with large surface areas will oscillate more than massive and spherical objects. An oscillating object will have a lower terminal velocity than a non-oscillating object. 30.5.3 Consequence of Dropped Object Impact The consequences of an impact are dependent on how a falling load actually hits the equipment (topside or subsea) or a structural component, i.e. velocity of the falling mass, hitting spot, impact angle, impact time, and contact area. Calculations are often made for ideal situations. It is often natural to distinguish falling loads between long cylindrical objects and bulky objects, because they have a different drop rate, trajectory/velocity in water, and effect on the structuxdequipment. Topside equipment such as pressure vessels, separators, are obviously vulnerable to the dropped object's impact. Subsea production systems and pipelines are also very sensitive to dropped objects. Some calculations have indicated that a falling load with a mass of 2 tons could easily damage an actuator on the subsea production system. The same loads applied to a pipeline may cause pipeline damage and leakage. For structural components, the following component parts are often of interest: a) Topside structure, b) Module support beams, c) Supporting structure, and d) Buoyancy compartments. Chapter 30 Risk Assessment.Applied to Offshore Structures 545 30.6 Case Study - Risk Assessment of Floating Production Systems 30.6.1 General A risk assessment may be conducted as part of the offshore field development and includes the following, All critical elements are appropriately selected and the corresponding performance standards are adequately defined for the life cycle of the FPS in terms of its functionality, availability, structural integrity, survivability, dependency and influence on the other critical elements. It should e demonstrated that the critical elements fit for purpose and meet the performance standards. Risk acceptance criteria are defined prior to the execution of risk assessment, and to provide a level of safety that is equivalent to that defined in the prescriptive rules and codes. All hazards with a potential to cause a major incident have been identified, their risks are evaluated and measures have been taken (or will be taken) to reduce the risk to the level that complies with the risk acceptance criteria. Type of risks for FPS depends on the type of vessel used and the geographical region it is sited. FPSOs used in the North Sea are mainly new vessels with turret system. The offloading tankers come to empty the storage tanks at frequency (approximately) once per week. The offloading tankers may represent a collision hazard to the FPSO with medium frequency and potentially high consequence. So far FPSOs in the west Afirica offshore are mainly based on spread mooring system and a single point mooring for oil export. FPSOs used in other geographical regions are mainly based on converted tankers. In the following, an FPSO Floating Production Storage and Offloading) for the Gulf of Mexico is chosen as an example to illustrate methods of risk assessment. The methods illustrated in this section may also be applied to other types of floating production systems such as TLPs, Spars and semi-submersibles. A risk assessment of FPSO may include evaluation of the following systems: Process Systems The process systems include, e.g.: e Process plant with three-stage separation, gas compression for export and gas turbine- driven power generation on deck piping, pressure vessels in production and storage facilities cargo tanks and crude pumping systems, offloading systems and its operation Process risk is mainly initiated by loss of hydrocarbons containment that might escalate to explosion and fire accidents. The risk assessment of process systems may be conducted using a conventional offshore QRA approach (Wolford, 2001), Development of isolatable sections Summarize the loss of containment frequency by using a parts count approach Identifylng spatial interactions that could lead to escalation 546 Pari V Risk Assessment Leak kquencies may be derived primarily hm generic databases that are available to the offshore industry. Emergency detection and process control response to a loss of containment event should be accounted for. API RP 14J (1993) has been used by the industry for the design and hazards analysis for facilities on offshore production installations. This RP mainly deals with the prevention of fire risk due to hydrocarbon ignition. Methodologies for hazard analysis are recommended. The API methodologies can be applied to assess explosion risk as well. Guidance is given on the risk management through platform equipment arrangement, hazard mitigation and personnel evacuation. Detailed check-lists are given in its appendix on facility layout (and emergency response/medical, escape and rescue), process equipment, safety and electrical systems, fire and gas leakage protection and mechanical systems, etc. Marine Systems The marine systems may include, e.g.: marine systems, such as cargo tanks, crude pump room, boilers and engine room, power generatiodsupply systems, ballast system and wing tanks, escape and evacuation system and equipment The risk assessment of marine systems is similar to that for process systems. The exception is the scope of marine system risk is broader than the loss of hydrocarbon containment. The majority of the marine system risk is fire due to fuel leakage and electrical systems. However, there is a lack of FPSO fire initiator frequency data for the appropriate quantification of fire risk. Structural Systems The structural systems may include, e.g.: 0 risers and flowlines topside structures helideck and helicopters operation flaresystem The structural system risk is covered in Part IV of this book. 30.6.2 Hazard Identification In an FPSO risk assessment, the primary objective of the hazard identification is to identify and register the hazardous events that may escalate into accidental events. The hazard identification task may be relatively coarse and subjective in the conceptual design phase, and become more specific in the detail design phase. A partial list of the typical hazards is given below. hull structures, especially the moonpool area that accommodates the turret if there is one position mooring systems, such as moorings and anchors, and/or dynamic position systems explosiondfires in cargo and ballast tanks The explosion and fire in cargo ballast tanks may result in hull structural failure and cause oil spill. explosions/fires in engine room and/or pump room Chapter 30 Risk Assessment Applied to Ofshore Structures 547 The explosions/fires in engine room and/or pump room may cause loss/delay of production, and escalate to cargo tanks. collisions from shuttle tanker or other vessels Shuttle tankers, supply vessels and pass-by vessels may collide into the FPSO due to failure of position mooring systems, errors in navigation or offloading operation, power failure etc. dropped objects Dropped objects may cause damage to structures leading to loss of buoyancy and cause damages to equipment and subsea flowlines leading to hydrocarbon leaks and personnel injuriedfatalities. extreme weather The weather conditions may be more severe than that considered in the design. Waves whose height is lower than the 100 year return design wave height but with more vibration sensitive wave periods may cause larger vessel motions and green water impacts. green water Green water can induce impacts loads on the forecastle, topsides along the deck edges of the vessel, and may cause damagelimpair of evacuation tunnels. structural failure such as corrosion defects and fatigue cracks Fatigue may be induced by wave loads and due to poor design of structural details. Corrosion defects may be found in cargo tanks, piping and pressure vessels. rupture in risers, flowlines and leaks in oftloading hose Failure of risers, flowlines and offloading hose may be caused by corrosion, fatigue and accidental loads. failure of station-keeping capacity A partial failure of the station-keeping system may lead to damages to risers resulting in gas leakage and fires. Loss of station keeping capacity may lead to collisions and grounding (in shallow water). 30.6.3 Risk Acceptance Criteria A risk matrix approach defined in Part V Chapter 29 may be used as the risk acceptance criteria and it consists of failure frequency and consequence. The failure frequency may be classed into high, medium, low and remote, each of them is defined below. High - an accident that occurred at least once in the past year and expected to occur again to the system, e.g. frequency > 0.1 Medium - an accident that might occur at least once in the life cycle of the system, no one would surprise if the accident occurs, e.g. O.Ol<fiequency<O.l Low - an accident is considered unlikely to occur. However, similar accidents have occurred once or twice in the industry worldwide, e.g. O.OOOl<frequency<0.01 Remote - an accident is credible, but not expected to occur in the life cycle of the system, e.g. frequency <0.0001. [...]... Platfonns”, D r h g Thesis, Division of Marine Structures, NTNU, MTAreport 19 9153 0 14 Karsan, D.I., Aggarwal, R.K., Nesje, J.D., Bhattachajee, S., Amey, C.E., Haire, B.M and Ballesio, J.E (1999), “Risk Assessment of Tanker Based Floating Production Storage and Offloading (FPSO) System in Deepwater Gulf of Mexico”, OTC 11000 15 Lassagne, M.G., Pang, D.X and Vieira, P (2001), “Prescriptive and Risk-Based... Petroleum Institute 3 Bai, Y and Pedersen, P Terndrup, “Elastic-Plastic Behavior of Offshore Steel Structures Under Earthquake Impact Loads”, International Journal of Impact Engineering, 13(1), pp 99-1 15 4 Burgan, B.A and Hamdan, F.H.(2002), “Response of Topside Structures to Fires and Explosions: Design Considerations”,Offshore Technology Conference, OTC 14130 5 CCPS (1999, “Chemical Transportation... and cleanup on board at sea can be difficult 5 62 Part V Risk Assessment Romer et a1 (1993) presented a risk assessment of marine transport of dangerous goods based on historical data that consist of 15 1 accidents in the period of 1986 to 1991 Their paper gave frequencies for various kinds of accidents, FN curves and frequencies and size of spills Loading Errors Improperly loaded cargo may adversely... of Risk Assessment within the Maritime Industry” EU RORO, MARIN, Wageningen, The Netherlands 14 USCG (1992), “The Marine Casualty Information Reporting Systems (CASMAIN), 1981-1991”, The U.S Coat Guard 15 Wennick, C.J (1993), “Collision and Grounding Risk Analysis for Ships Navigating in Confined Waters”, Journal of Navigation, Vol 45 (l), pp 80-91 16 Yoshida, K et a1 ( 0 0 , “Risk Assessment”, Proceedings... the main activities, duration, and the cost in each phase is shown in Figure 32.1 EXPLORATION TIME DEVELOPMENT OPERATION GRANTING OF PERMIT 1 2 3 4 5 6 7 8 Drilling 9 10 11 Wells Simulations 12 13 14 15 16 17 18 19 20 21 22 23 24 2s 26 DECOMMISSION Share of Technical Cost Figure 32.1 10 to 20Yo 40 to 60% Field Development Phases The exploration phase starts after the field license is awarded If an . systems, fire and gas leakage protection and mechanical systems, etc. Marine Systems The marine systems may include, e.g.: marine systems, such as cargo tanks, crude pump room, boilers and. assessment of marine systems is similar to that for process systems. The exception is the scope of marine system risk is broader than the loss of hydrocarbon containment. The majority of the marine. of fire risk. Structural Systems The structural systems may include, e.g.: 0 risers and flowlines topside structures helideck and helicopters operation flaresystem The structural system