Post-damage (Electrical) Systems Availability

Transcription

1 Post-damage (Electrical) Systems Availability! Konstantinos Sfakianakis MEng, PhD candidate! University of Strathclyde Department of Naval Architecture, Ocean and Marine Engineering

2 ! Design for Safety / Probabilistic-based Design

3 Presentation Outline Introduction Approach Adopted Example Case study (1) Case study (2) Concluding Remarks

4 Introduction

5

6 Containing Risk Today (Human Life) SOLAS!! Consensus-based, minimum standards of safety Targeting to reduce consequences Historical risk, reflecting specific data sets! Compliance with Rules/Regulations

7 Introduction Passenger ships accidents

8 Introduction Accident Statistics

9 Introduction Even though: Ship systems design is addressed as one of the most dominant issues in shipping industry Systems reliability is considered as driving force during the systems design

10 Introduction Up to now The strict international rules and regulations concerning systems safety No room for innovations! The systems design simplicity based on past experience with lack of a systematic and structured approach

11 Introduction Up to now In specific: SOLAS 90 and independent classification societies imposing the systems components location, geometry and in some cases topology, to comply with their requirements in order to be accepted without the presence of any alternative design.

12 Introduction Example- SOLAS 90 Reg. 41- Main source of electrical power and lighting systems! Par. 3. The main switchboard shall be placed relative to one main generating station An environmental enclosure for the main switchboard, such as may be provided by a machinery control room situated within the main boundaries of the space, is not to be considered as separating the switchboard from the generators.! Par. 4. Where the total installed electrical power of the main generating sets is in excess of 3MW, the main busbars shall be subdivided into at least two parts which shall normally be removable links or other approved means;

13 Introduction Up to now SOLAS 90 The previous example highlights the deterministic approaches based on past experience that the ship systems are designed and reveals the need for increased designers flexibility adopting innovative design methodologies.! The absence of the nature of the accident lead to insufficient, in terms of safety, and possibly more expensive systems designs.

14 Introduction New international regulations - Safe Return to Port (1) New international regulations (SRtP) address the availability issues for more efficient and safe systems design.! Intend to ensure high systems reliability (not only in normal operation but also in emergencies)! Consider abandonment as the last safety barrier to be used when all means of saving the vessel failed! Impose capability by the vessel to sail to the nearest port post-casualty (3 hours operation)

15 Introduction IMO (SLF 47/48) Passenger Ship Safety

16 Introduction Example of Loss Scenario Flooding / Collision prevention mitigation

17 Introduction Systems Availability Analysis SRtP Casualty Scenarios

18 Introduction Safe Return to Port (2) A passenger ship shall be designed so that it s safety-critical systems remain operational when the ship is subject to flooding of any single watertight compartment (SOLAS II-1/8-1)! Design criteria for (safety-critical) systems to remain operational after a fire casualty (SOLAS II-2/22)

19 Introduction Safe Return to Port Casualty Threshold Casualty threshold, in the context of a fire, includes: Loss of space of origin up to the nearest A class boundaries, which may be a part of the space of origin, if the space of origin is protected by a fixed fire-extinguishing system; or Loss of the space of origin and adjacent spaces up to the nearest A class boundaries which are not part of the space of origin.

20 Introduction Systems Availability Analysis SRtP Safety Critical Systems

21 Introduction Safe Return to Port Safe Areas Safe areas functional requirements: The safe area(s) shall generally be an internal spaces(s) The safe areas shall provide all occupants with the following basic services to ensure that the health of passengers and crew is maintained: Sanitation Water Food Alternate space for medical care Shelter from the weather Means of preventing heat stress and hypothermia Light Ventilation

22 Introduction Safe Return to Port Ventilation design shall reduce the risk of smoke and hot gases that could affect the use of the safe area(s)! Means of access to life-saving appliances shall be provided from each are identified or used as a safe area, taking into account that a main vertical zone may not be available for internal transit.

23 Introduction Safe Return to Port (4) Flexibility to designers SRtP afford design flexibility through the Alternative Design and Arrangements (AD&A) framework (SOLAS II-1/55) through the demonstration of equivalent designs.

24 Approach Adopted

25 Introduction Up to now Systems Reliability Concerning the qualitative and quantitative systems reliability analysis, tools as: Fault Tree Analysis (FTA) Failure Mode and Effects Analysis (FMEA) Failure Mode, Effects and Criticality Analysis (FMECA) HiP-HOPS

26 Approach Adopted Practice Today Driven by the absence of post-casualty systems performance in past regulations: Reliability analysis based on components failure description and their topology (FMEA/FMECA, FTA) Independent to the ship environment! Past operational experience

27

28 Approach Adopted Practice Today - Example

29 Approach Adopted Practice Today - Example In the example the top event no electrical power at motor 1 occurs if one of the following failure modes takes place (ordered from consumer to generator):! Interruption (connection failure) in cable1, Short circuit in cable1, Short circuit in switch1, switch1 inadvertently open, Short circuit in switchboard1, Short circuit in genswitch1, genswitch1 inadvertently open.

30 Approach Adopted Practice Today - Example The failure probability of all components within a time span t and for a failure rate λ is calculated using the cumulative exponential failure description:!!!!! The system failure probability is calculated to for t = 24 hours

31 Approach Adopted Practice Today - Example

32 Approach Adopted Adopting a performance-based approach The new regulations implicitly introduce concept of the absolute survivability, i.e. floatability combined with the survivability of the functions. Therefore, it is thought that the probabilistic approach used in the damage stability to evaluate safety of the ship (index-a) can be successfully used for systems availability assessment. The probabilistic, index-based, methodology not only allows meet the criteria but it additionally ensures consistency between survivability of vessel and onboard systems.

33 Approach Adopted Adopting a performance-based approach Performance-based approach focused on the logical and topological modelling of the electrical distribution energy systems into the ship environment Systems quantitative performance assessment at emergencies under statistical flooding damages

34 Approach Adopted Logical and Topological Modelling Mapping the functionality of each system to logical/dependency structures These structures comprises physical, functional and spatial distribution relations of each system

35 Approach Adopted System dependency tree example of propulsion system

36 Approach Adopted Damage scenarios - Initiation of the systems and components failures at emergencies Use of collision and grounding casualties Damage scenarios are generated with use of NAPA software The probability of damage occurrence at each compartment and at all possible combinations of compartments are called p-factors. The p-factors for all damage cases along with a list of compartments being damaged are passed as input for probabilistic assessment of systems availability.

37 Approach Adopted Availability Analysis After the components placement within the vessel environment, with the parallel creation of dependency structures, the computational stage is following. All the system components and functions are subjected to individual damage scenarios, which are defined as a combination of probability of occurrence p i and a list of spaces (rooms) being affected. An average probability of system being unavailable given any considered damage scenario has happened is evaluated

38 Approach Adopted Availability Analysis The element p i denotes probability of the i-th scenario and F j stands for aggregated probability of j-th system being 1 n " # unavailable. The normalising factor $ pi % is used as p-factors should be sum up to one in case of all possible & i= 1 ' damages are considered. However, for the case of SRtP compliance, only damages occurring in any single WT compartment are assessed and so the normalising factor equals the sum of the probability of damage occurrence in each specific WT compartment. Furthermore, the precise meaning of the components F j of the vector F (probability of j-th system being unavailable) is an average probability of system being unavailable given any considered damage scenario has happened

39

40 Example

41 Example Apply the electric power system to the ship environment (with safety constraints) Example with the use of isys software 2 decks of a notional ship! The main switchboard supplying two loads is defined! The cable routing is defined! Application of flooding damage is under investigation

42 Example (explanation of isys software) The diagram of the electrical components is presented in the next slide with their corresponding topologies! The Switchboard-Load1 are connected directly since they are placed in rooms one next to the other. However, the connection of Switchboard-Load2 is achieved through the cable routing of Load1 room.! Afterwards, we apply a damage to each room and the results show the damage effects in our systems. First, we apply to room of load2, after to the load1 and last at the switchboards room.

43 load1 switchboard load2

44 load2 load1 switchboard

45 Example We can easily notice that when a damage applied to the room of Load1 (middle compartment), there isn t supply to Load2.! By re-routing the distribution system and adding circuit-breakers for elimination of the damage, we can achieve the appropriate redundancy of power distribution. Having this in mind, the redundant energy routing is through the upper deck.! We are now add a ship function to supply power the load2 either through the initial path or through the new design.

46 New design of power distribution

47 Example - Results As we see below, during the damage in the middle compartment, the initial cabling is damaged and eliminated by the circuit-breakers to both directions, as well as the Load2 is supplied through the new design path

48 Case Study(1) SAFEDOR

49 Alternative design

50 Results Explanation Average probability of systems being unavailable given collision and flooding within a single WT-compartment.! Average probability of systems being unavailable given uncontrolled fired on a single deck within a single MVZ

51 Results

52 Case Study(2)

53 Case Study Compare a conventional with an alternative design of a RoPax ship electrical distribution systems considering: Damage scenarios to any single WT compartment All possible damage scenarios Damage propagation is not investigated Rooms and spaces current usability is not taken under consideration

54 System Propulsion Steering Bilge & Ballast Emergency Sub- System Ventilation Number of Location Source Component sets (in use) Engine Room Supply fan 2(1) Upper deck MSWBD Auxiliary Room Supply fan 2(1) Upper deck MSWBD Engine Room Exhaust fan 1(1) 4 th deck MSWBD Auxiliary Room Exhaust fan 1(1) 4 th deck MSWBD Air M/E Air Compressor 2(1) Engine Room SWBD_4 Fuel Lubrication ME FO Booster pump 2(2) Engine Room SWBD_3 Auxiliary SWBD_1 DE FO Booster pump 2(2) Room Auxiliary SWBD_1 DO Transfer pump 1(1) Room Auxiliary SWBD_1 FO Transfer pump 1(1) Room Auxiliary SWBD_1 DO Purifier 1(1) Room FO Purifier 2(2) Auxiliary Room SWBD_1 Engine Room SWBD_3, LO pump 4(2) SWBD_5 Engine Room SWBD_4, Reduction Gear LO pump 4(4) SWBD_5 Auxiliary SWBD_2 LO Purifier 2(2) Room Auxiliary SWBD_2 DE LO Purifier 1(1) Room Engine Room SWBD_5, ME LT FW cooling pump 2(2) SWBD_6 ME FW cooling pump 2(2) Engine Room SWBD_3 Cooling ME HT FW cooling pump 2(2) Engine Room SWBD_3 Engine Room SWBD_3, SW cooling pump 3(2) SWBD_6 Auxiliary SWBD_2 DE LT FW cooling pump 2(2) Room Engine Room SWBD_4, SWBD_6 CPP CPP Propeller pitch setting pump 4(4) Rudder Steering Gear 4(2) Steering Room MSWBD Bow Thruster MSWBD Thruster Bow Thruster 1(1) Room Stern Thruster MSWBD Stern Thruster 1(1) Room Bilge, Fire & Ballast pump 2(1) Engine Room SWBD_4 Auxiliary SWBD_2 E/R Fire & Bilge pump 1(1) Room Auxiliary SWBD_1, Ballast pump 2(1) Room SWBD_2 Drencher 1(1) Engine Room MSWBD Auxiliary Room Supply fan 1 Upper deck EMSWBD Engine Room Supply fan 1 Upper deck EMSWBD M/E Air Compressor 1 Engine Room EMSWBD Steering Gear 2 Steering Room EMSWBD Bilge, Fire & Ballast pump 1 Engine Room EMSWBD

55 Passenger vessel WT arrangements

56 Case Study Conventional design

57 Case Study Alternative design

58 Case Study Results System Conventional Alternative Sub-systems Conventional Alternative Ventilation Air Propulsion Fuel Lubrication Cooling CPP Steering Rudder Thruster Average probability of systems being unavailable given collision and flooding within any single WT compartment. Bilge & Ballast Emergency System Conventional Alternative Sub-systems Conventional Alternative Ventilation Air Propulsion Fuel Lubrication Cooling CPP Steering Rudder Thruster Average probability of systems being unavailable given collision and flooding considering all the damage scenarios Bilge & Ballast Emergency

59 Concluding Remarks (1) Considering Safety as a design objective, more efficient in terms of costs, spaces and safety, electrical onboard distribution energy systems can be obtained The redundant components can be minimised and the option of redundant flow paths, usually close to the centre of the ship and in WT compartments far from the MSWBD through higher decks, is possible

60 Concluding Remarks (2) The size of the emergency source of power can be optimised Taken under consideration all possible damage scenarios during the assessment, more accurate design recommendations can be applied. Approval of Alternatives and Equivalents (IMO MSC/Circ. 1455)

61 Concluding Remarks (3) This work is a part of the holistic performance-based approach for the design of the electrical onboard energy systems aiming to the multi-objective optimisation in terms of energy efficiency, safety and cost.

62 ! Design for Energy Efficiency / Dynamic Energy Modelling

63 Presentation Outline Introduction Approach Adopted Example Cargo Design of Power Generation (Capacity) Power Management Systems (PMS) PMS during Design Example PMS during Operation - Example Multi-objective optimisation

64 Introduction

65 Legislation Environmental performance of ships Energy efficiency EEDI, EEOI, SEEMP Environmental performance

66 Legislation Energy Efficiency Design Index (EEDI) CO 2 emissions [grams CO 2 ] Benefit to society [tonnes x nm]

67 Legislation Energy Efficiency Design Index (EEDI) Design index applicable to new ships only Aimed to stimulate more efficient ship designs and technologies Regulated through IMO as technical means to control CO 2 New ships must meet a required efficiency level (reference line)

68 What do the regulations say? Goal-based legislation (performance standard) MARPOL Annex VI, Chapter 4, Regulations on Energy Efficiency for Ships Enters into force 1 st January 2013 EEDI (new ships) SEEMP (all ships) Both will form part of the International Energy Efficiency Certificate (IEEC) EEOI: optional tool for operational indexing

69 Where do the regulations apply? Targeted ship types Ships > 400 GT (excluding ships with gas turbine, diesel-electric and hybrid propulsion): Bulk carrier Gas carrier Tanker Container ship General cargo ship Refrigerated cargo carrier Combination carrier Passenger ship 1 Ro-Ro ships 1 (cargo, vehicle, passenger)! 1 EEDI to be calculated but not yet subject to regulatory limits

70 EEDI How will it be implemented? A 4-phase implementation approach 0% -10% -15% -20% -30% Phase 0: Cut off limit Capacity [DWT or GT] Phase 1: Phase 2: Phase 3:

71 World Fleet (fuel-saving market) Share by EEDI-targeted ships

72 World Market EEDI targets Ship Type Medium (500<GT<25000) Large (25000<GT<60000) Very Large (GT>60000) Total Bulk Carriers % % % % Gas Tankers 995 3% 192 2% 319 8% % Oil and Chemical Tankers % % % % Container Ships % % % % General Cargo Ships % 228 3% 0 0% % Passenger Ships % 268 3% 130 3% % Ro- Ro Cargo Ships 843 3% 559 6% 126 3% % Total % % % % 55% of the world market

73 Approach Adopted

74 The Way Forward Dynamic Energy Modelling Modelling the dynamics of energy flows within complex engineering systems as function of time Accurate assessment of life-cycle fuel costs and carbon footprint early in the design stage and during operation Design for energy efficiency and minimum environmental impact, alongside other design objectives

75 Dynamic Energy Modelling Concept DEM integrates knowledge from component-level to ship-system level The performance of ship systems is assessed by time-domain simulation revealing the true energy performance of systems DEM takes into consideration inherent properties of systems in relation to environmental conditions and the operational profile of the ship.

76 Dynamic Energy Modelling From Static Balance to Time-Domain Simulations Response Function Methods Based on first principles Accommodate relatively complex systems Reasonable for short time periods System parameters linear & time invariant, hence results may be unrealistic Complex Systems Numerical Simulation Methods Founded on first principles Accommodate design and operational complexity Account for complex interactions Energy flow instead of energy balance Suitable for any design stage No limit in time periods (seconds to life cycle) Steady State Systems Dynamic Systems Static Balance Methods Accommodate isolated systems No mechanism for complex interactions Safeguarding against worse-case scenarios Simple Dynamic Methods Based on regression models of past / existing designs or simplified simulations Can not accommodate innovative designs Simple Systems

77 Dynamic Energy Modelling Energy Systems and Components Components Systems Global System/Ship i Sum Sum

78 Dynamic Energy Modelling Integration Superstructure Component Auxiliary Energy Added Resistance Electric Power System Wave Resistance Engine Room Systems Propeller Prime Mover Component Advanced Surface technology Frictional resistance/ Hull coatings Time-domain simulation of power demand Optimisation

79 Dynamic Energy Modelling Methodology Global Efficiency! Local Efficiency! Operational Costs! Life-cycle Performance! Other

80 Dynamic Energy Modelling Approach Adopted ELECTRIC ENERGY MECHANICAL ENERGY CHEMICAL ENERGY DEM PLATFORM Integration / Simulation Optimisation THERMAL ENERGY LIFE-CYCLE ENERGY MANAGEMENT DESIGN OPERATION RETROFITTING

81 Dynamic Energy Modelling Electric Power System - Objectives Scope Electric power components (generators, motors, transformers etc) Control strategies during operation Objectives Dynamic modelling of shipboard electric power system Power Management System (PMS) to monitor and control the overall performance of the marine power system.

82 Dynamic Energy Modelling Electric Power System Practice today The sizing of the diesel generators is based on static balance calculations ( i.e. electric load analysis) Based on the maximum values of these calculations, appropriate engines are selected from suitable database All auxiliary systems connected to the engines are sized based on their maximum power output! Size and number of diesel generators Maximum power during operation

83 Dynamic Energy Modelling Electric Power System What we do Integrate all electric components into an electric power system (manufacturers data and first-principles models) Identification of dominant parameters affecting the process Create knowledge intensive models (KIMs) to capture all the essential information at component level Integrate the electric power system into the ship system model

84 Dynamic Energy Modelling Electric Power System Link with other systems The electric power system is linked with (almost) all other energy systems onboard ships For any changes in the power demand at the boundaries of the system (on/off of motors, lighting, etc) the system balances at a different operating state Interaction with other energy systems Interaction with Environment Interaction with Diesel Engine Module Interaction with other energy systems

85 ! Example Cargo ship

86 Example On-board electric power system of two bulk carriers. Example Cargo bulk carrier ship 85,400 GT! Diesel Generator x3 : 750kVA ( 600kW) Induction motors : kw Transformers: 15kVA, 90kVA Lighting : 15kVA, 100kVA Minor equipment: 35 kw It is investigating at sea going and as a result only one Diesel Generator is in function. The numbers next to the components are the number of components that are in function at the on-board measurements.

87 Diagram BUS1 Three%phase Induction1 Motor MCTC Pump x40 Diesel1 Engine Three%phase1 Synchronous1 Generator Three%phase1 Induction1 Motor MCTC Fan x12 Three%phase1 Transformer Lighting MCTC Minor1 Equipment MCTC

88 Results (Active & Reactive Power) My results On-board measurements

89 Results (Power Factor) My results On-board measurements

90 Design of Power Generation (Capacity)

91 Design of Power Generation Capacity - Empirical formulae Empirical formulae can be used successfully to obtain a first estimation if the electric power demand in the pre-design stage, if the formulas are based on a sufficient number of ships with the same mission statement and comparable size. However, for the detailed design of the ship and electrical systems one of the next methods is indispensable to get a more reliable result.! When empirical formulae are at hand, they can be used to determine the electric power demand or installed electric power by using, for instance, the main dimensions of the ship such as deadweight size or installed propulsion power. A common formulae is the one below that uses the installed propulsion power to determine the electric power demand as sea for a conventional cargo vessel without special equipment such as cargo refrigeration system or bow thruster. As a rule of thumb, the electric load when manoeuvring is 130% of the electric load at sea, and the load in power is 30 % to 40%.

92 Empirical formulae! P EL = *( P ) MCR! The use of empirical formulas might be successful provided that the ships are comparable in size and mission. 0.7

93 Electric Load Analysis The most widely method for determining the electric power demand is the so-called electric load analysis or electric load balance. The balance sheet lists all electric power consumers, highlighting the nominal properties of all the electric consumers. (predefined_ operational conditions of the ship). In addition, estimating a load factor and a simultaneity factor for all consumers at each operating state, the capacity of the generated power, in turn the size of the power generation units, can be approximated. However, the estimation of the load and simultaneity factors is the most difficult part of the electric load analysis. The load factor indicates the relative (%) load of the machinery and thus specifies how much electric power is absorbed in an actual situation. A steering gear pump, for example, will only occasionally be fully loaded, so a typical load factor for a steering gear pump is 0.1. The simultaneity factor accounts for pieces of machinery that are not operated continuously but intermittently. The simultaneity factor indicates the relative (%) mean operational time of the machinery. It is often possible to make a good estimation of this factor by comparing the machine capacity and the average capacity demand. Both factors vary between 0 and 1. Usually, no distinction is made between the load factor and the simultaneity factor, and the two factors are combined into one service factor. This does, not provide a clear insight into the actual load demand.

94 Electric Load Analysis To this end, both factors are often estimated too high, in order to minimise the risk of designing a plant with a generator capacity that is too small. This results in an overestimation of the electric power demand, and consequently the chosen generator capacity is too large. The drawbacks out of this estimations are the high initial investment, the operation of the diesel generator sets away from their optimum point increasing the fuel consumption and the pollutants and finally the increase of the maintenance costs.

95 Electric Load Analysis

96 Power Management Systems (PMS)

97 Power Management Systems (PMS) The PMS is very important during all the ship stages: design, retrofitting and operation. At the design stage PMS as an optimiser can be useful for the sizing and the number of the power generation units that has to be online at each operating stage based on the fuel consumption. During operation the PMS can be helpful for fuel optimisation, disturbance rejection and in the maintenance of the equipment of generation, conversion, distribution and consumption through the actions of: load sharing and unit commitment, load shedding and the quality of power that it supplies. At the retrofitting the re-configuration of the existing control system in conjunction with the appropriate location of monitoring and data collection points, can be obtained through the PMS improving so the energy efficiency as the marine power systems reliability

98 Power Management Systems (PMS) A sophisticated ship power management system usually provides the following main functions: Diesel generator (DG) start, stop control Auto-synchronizing of generators and breaker control Load depend start, stop Unit commitment - Load sharing Load increase control Blackout monitoring Load shedding Shaft generator (SG) load transfer

99 PMS during Design - Example

100 PMS during Design - Example Number of installed units is selected, Ng=4; wr,gi = Pr,gi/Prg is manually pre-selected and listed in the table; Total installed power Prg = kw.

101

102 PMS during Operation - Example

103 Optimisation constraints (1) The optimization of the operation of the ship power system is imposed to several constraints and limitations which are briefly described next. There are several technical constraints to be applied in order to ensure system safe operation as well as physical rules to be followed.! Power balance constraint. It assures balance between generation and consumption as well as frequency stability. High loading constraint. Generator should not be loaded above a certain power level for more than a specific time interval as thermal and mechanical losses are increased and blackout prevention capability is limited. Low load constraint (technical minimum). The engine should not be loaded below a certain value specified by the engine manufacturer in order to reduce the maintenance costs and possible damage.

104 Optimisation constraints (2) GHG emissions constraint. EEOI should be monitored online and limited below a certain upper limit. Ramp rate constraint. High rate of change of generator loading must be avoided in order to eliminate mechanical stress and damages. Blackout prevention constraint. It defines the maximum allowable continuous loading of the generators where the system is blackout-proof. Generator start/stop constraint. Frequent start/stop of the generator results in increased maintenance cost and fuel consumption. It is a secondary priority constraint, and it can be applied by imposing a time window between successive start/stop of the generator.

105 Operation

106

107 Multi-Objective Optimisation

108 From Spiral Design to Multi-Objective Design

109 Multi-Objective Design Optimisation Basic Principles Definitions Constraints Design parameter D Design space Objective = max Area(D) Two-objectives problem Three-objectives problem Four-objectives problem...

110 Multi-Objective Design Optimisation Optimisation Strategies Two-Objectives Problem: maximise areas of two circles within a rectangle Strategies: Sequential: subdivide problem into phases maximise first, then second Holistic optimisation: maximise both simultaneously

111 Multi-Objective Design Optimisation Optimisation Strategies (Sequential / Stepwise) Phase 1:! max Phase 2: Alternative 1 Alternative 2! max! max max max Number of overall alternatives is strongly limited Design space exploration is inherently limited

112 Multi-Objective Design Optimisation Optimisation Strategies (Holistic)! max max! max max max max max max max max max! max Number of overall alternatives is limited only by computational resources (unlimited!) Design space is well explored maximum effect!

113 Multi-Objective Optimisation Multiple Conflicting Objectives (KPIs) Common Key Design Parameters - reason for conflict! Design Space Key Design Parameter Obj Design Objective/Goal

114

115 Thank you!