III5 1 Dependent failure modelling

50 %
50 %
Information about III5 1 Dependent failure modelling
Entertainment

Published on December 30, 2007

Author: Gavril

Source: authorstream.com

Basic Level 1. PSA course for analysts:  Basic Level 1. PSA course for analysts Dependent failure analysis IAEA Training in level 1 PSA and PSA applications Content:  Content Types of dependencies in the PSA Physical dependencies Functional dependencies Location/environmental dependencies Data based dependencies Plant configuration related dependencies Common Cause Failures Human dependencies PHYSICAL DEPENDENCIES:  PHYSICAL DEPENDENCIES EXAMPLES COMMON SUCTION VALVE FOR TWO PUMPS AC POWER SUPPLY DC POWER SUPPLY INTERLOCKS FOR PUMPS, VALVES, CIRCUIT BREAKERS AUTOMATIC START / ALIGNMENT SIGNALS RESTART SIGNALS COOLING WATER TREATMENT DEPENDENCY MATRICES EVENT TREE / FAULT TREE LOGIC STRUCTURE EVENT TREE / FAULT TREE LINKING FUNCTIONAL DEPENDENCIES:  FUNCTIONAL DEPENDENCIES EXAMPLES INJECTION CRITERIA FOR LOCAs INJECTION REQUIRED FOR RECIRCULATION MAKEUP AND STEAM RELIEF REQUIRED FOR SECONDARY HEAT REMOVAL OPERATOR ACTIONS TO START / ALIGN EQUIPMENT TIMING OF FAILURES AND RECOVERY ACTIONS COORDINATED OUTAGES AND PLANNED MAINTENANCE COOLING WATER HVAC AND ROOM COOLING TREATMENT EVENT SEQUENCE DIAGRAMS SUCCESS CRITERIA FOR SYSTEMS AND OPERATOR ACTIONS TIME INTEGRALS FOR FAILURES AND RECOVERY ACTIONS PLANNED MAINTENANCE MODELS EVENT TREE / FAULT TREE LOGIC STRUCTURE EVENT TREE / FAULT TREE LINKING LOCATION / ENVIRONMENTAL DEPENDENCIES:  LOCATION / ENVIRONMENTAL DEPENDENCIES EXAMPLES STRUCTURAL FAILURES / SEISMIC EVENTS FIRES FLOODING TURBINE MISSILES WATER SPRAY HVAC AND ROOM COOLING INTAKE PLUGGING TREATMENT SPATIAL INTERACTIONS ANALYSES EXTERNAL EVENTS ANALYSES EVENT TREE / FAULT TREE LOGIC STRUCTURE EVENT TREE / FAULT TREE LINKING DATA-BASED DEPENDENCIES:  DATA-BASED DEPENDENCIES EXAMPLES MULTIPLE COMPONENT MAINTENANCE COORDINATED OUTAGES AND PLANNED MAINTENANCE OBSERVED COMBINED FAILURE RATE FOR MULTIPLE SIMILAR COMPONENTS IS HIGHER THAN THE INDEPENDENT PRODUCT OF THE SINGLE COMPONENT FAILURE RATES TREATMENT COMMON MAINTENANCE BASIC EVENTS PLANNED MAINTENANCE MODELS COMMON CAUSE FAILURE BASIC EVENTS COMMON CAUSE FAILURE PARAMETERS EVENT TREE / FAULT TREE LOGIC STRUCTURE EVENT TREE / FAULT TREE LINKING HUMAN DEPENDENCIES:  HUMAN DEPENDENCIES EXAMPLES TIME WINDOW FOR OPERATOR RESPONSE SIMILAR FUNCTIONS MULTIPLE OPTIONS / PRIORITIES PROCEDURES / TRAINING PERSONNEL / STAFFING LOCATION PRECEDING SYSTEM SUCCESSES / FAILURES PRECEDING OPERATOR SUCCESSES / FAILURES TREATMENT ORGANIZE MODELS TO DISPLAY OPERATOR ACTIONS THERMAL / HYDRAULIC ANALYSES FOR TIME WINDOWS COGNITIVE RESPONSE / IMPLEMENTATION TASKS EVENT TREE / FAULT TREE LOGIC STRUCTURE EVENT TREE / FAULT TREE LINKING PLANT CONFIGURATIONS PLANT OPERATING ALIGNMENTS:  PLANT CONFIGURATIONS PLANT OPERATING ALIGNMENTS VERY IMPORTANT FOR SHUTDOWN PSA OPERATIONAL CONSIDERATIONS PRESSURE, TEMPERATURE, COOLING MODE RCS AND CONTAINMENT STATUS SYSTEM ALIGNMENTS / SIGNALS / ISOLATION TESTING / MAINTENANCE CONSIDERATIONS PLANNED MAINTENANCE TESTING PROGRAMS PLANNED MAINTENANCE ALIGNMENTS MAY ALSO APPLY FOR FULL-POWER PSA COMMON CAUSE FAILURES COMMON CAUSE FAILURE GROUPS:  COMMON CAUSE FAILURES COMMON CAUSE FAILURE GROUPS SIMILAR COMPONENTS SAME FAILURE MODES SIMILAR OPERATING DUTY CYCLES SIMILAR TESTING, INSPECTION, MAINTENANCE MAY APPLY ACROSS DIFFERENT SYSTEMS COMMON CAUSE FAILURES LEVEL OF DETAIL:  COMMON CAUSE FAILURES LEVEL OF DETAIL BETA-FACTOR MODEL SIMPLE NUMERICALLY CONSERVATIVE REASONABLE DATA FOR MANY COMPONENTS AND FAILURE MODES MULTIPLE GREEK LETTER (MGL) MODEL MORE COMPLEX FAULT TREES / CUTSETS LOGICALLY MORE CORRECT NUMERICALLY MORE REALISTIC VERY SPARSE DATA FOR MORE THAN 3 FAILURES COMMON CAUSE FAILURES LEVEL OF DETAIL (cont.):  COMMON CAUSE FAILURES LEVEL OF DETAIL (cont.) OTHER PARAMETRIC MODELS LIMITED BY SAME DATA “LETHAL SHOCKS” AFFECT ALL COMPONENTS IN A GROUP VERY LIKELY TO BE CAUSE FOR MORE THAN 3 OR 4 CORRELATED FAILURES MODELS FOR LARGE NUMBERS OF COMPONENTS LARGE NUMBER OF COMBINATIONS IN POPULATION BEWARE OF FUNCTIONAL IMPACTS FROM SPECIFIC COMBINATIONS COMMON CAUSE FAILURES EXAMPLE - 10 RELAYS:  COMMON CAUSE FAILURES EXAMPLE - 10 RELAYS RELAY FAILURE IMPACTS TRAIN A: RELAYS RA1 * RA2 TRAIN B: RELAYS RB1 * RB2 TRAINS A * B: RELAYS RA1 * RB1 * RXX COMBINATIONS IN POPULATION 2 RELAYS: (10!) / (8!*2!) = 45 3 RELAYS: (10!) / (7!*3!) = 120 4 RELAYS: (10!) / (6!*4!) = 210 COMMON CAUSE FAILURES EXAMPLE - 10 RELAYS (cont.):  COMMON CAUSE FAILURES EXAMPLE - 10 RELAYS (cont.) TRAIN A: 1 / 45 OF DOUBLE FAILURES + 8 / 120 OF TRIPLE FAILURES TRAIN B: 1 / 45 OF DOUBLE FAILURES + 8 / 120 OF TRIPLE FAILURES TRAINS A * B: 1 / 120 OF TRIPLE FAILURES COMPLETE MGL EXPANSION MORE REALISTIC MODELS AND RESULTS MAY BE WORTHWHILE EVEN IF FAILURE OF 4 OR MORE RELAYS CAUSES SEVERE CONSEQUENCES COMMON CAUSE FAILURES COMMON CAUSE DATA SCREENING:  COMMON CAUSE FAILURES COMMON CAUSE DATA SCREENING TABULATED PARAMETER VALUES BROAD APPLICABILITY AUTHORS’ JUDGMENT MAY NOT BE CONSERVATIVE FOR ALL APPLICATIONS ACTUAL EVENT SUMMARIES MOST USEFUL EVENT REVIEW / SCREENING DO NOT USE “BETTER” TRAINING, PROCEDURES, PEOPLE, ETC. AS BASIS FOR REMOVING EVENTS POSSIBLE MORE SEVERE PLANT-SPECIFIC IMPACTS THAN AT OCCURRENCE PLANT HUMAN DEPENDENCIES ELEMENTS OF A HUMAN ACTION:  HUMAN DEPENDENCIES ELEMENTS OF A HUMAN ACTION IDENTIFICATION DIAGNOSIS COGNITIVE DECISION RESPONSE IMPLEMENTATION } HUMAN DEPENDENCIES HUMAN BEINGS ARE NOT HARDWARE:  HUMAN DEPENDENCIES HUMAN BEINGS ARE NOT HARDWARE HUMAN RELIABILITY CANNOT BE EVALUATED OUT OF CONTEXT HUMAN PERFORMANCE DEPENDS ON THE ENTIRE HISTORY OF ACCUMULATED KNOWLEDGE, EXPERIENCE, TRAINING, GUIDANCE, AND INFORMATION UNTIL THE TIME OF RESPONSE HUMANS INTERPRET THE INFORMATION THAT THEY RECEIVE, EVALUATE ITS RELEVANCE AND MEANING, FORM A CONCLUSION, AND RESPOND INFORMATION, INTERPRETATION, AND DECISION DEPEND ON CONTEXT HUMAN DEPENDENCIES SCENARIO-BASED PERSPECTIVE:  HUMAN DEPENDENCIES SCENARIO-BASED PERSPECTIVE PSA MODELS CONTAIN VERY LARGE NUMBERS OF INDIVIDUAL SCENARIOS (“SEQUENCES”, “CUTSETS”, ETC.) IDENTIFY IMPORTANT DIFFERENCES THAT AFFECT HUMAN RESPONSE GROUP SCENARIOS AND DEFINE PSA ACTIONS BASED ON SUCCESS CRITERIA AND BOUNDARY CONDITIONS FOR HUMAN PERFORMANCE MANUAL START OF STANDBY EQUIPMENT IS A SCENARIO-BASED COGNITIVE ACTION HUMAN DEPENDENCIES SCENARIO-BASED CONSIDERATIONS:  HUMAN DEPENDENCIES SCENARIO-BASED CONSIDERATIONS INITIATING EVENT AVAILABLE TIME WINDOW AVAILABLE EQUIPMENT CUES, INDICATIONS, AND ALARMS PROCEDURES, TRAINING, AND EXPERIENCE COMPETING PRIORITIES PREVIOUS OPERATOR ACTIONS (SUCCESSES AND FAILURES) HUMAN DEPENDENCIES IDENTIFY / DISPLAY DEPENDENCIES:  HUMAN DEPENDENCIES IDENTIFY / DISPLAY DEPENDENCIES PSA MODELS SHOULD: DISPLAY OPERATOR ACTIONS IN SCENARIO CONTEXT IDENTIFY ALL CONDITIONS WHERE OPERATOR ACTIONS ARE COMBINED THROUGH “AND” LOGIC DIFFICULT TO IDENTIFY SCENARIO CONTEXT AND COMBINED ACTIONS IN FAULT TREE FORMAT USUALLY REQUIRES MODEL SOLUTION NUMERICAL VALUES MAY SUPPRESS CUTSETS “SCREENING VALUES” MAY BE OPTIMISTIC EVENT TREE FORMAT GENERALLY BETTER DEFINES SCENARIO CONTEXT IDENTIFIES COMBINED ACTIONS HUMAN DEPENDENCIES FACTORS THAT REDUCE HUMAN DEPENDENCE:  HUMAN DEPENDENCIES FACTORS THAT REDUCE HUMAN DEPENDENCE PRECEDING OPERATOR SUCCESS LONG TIME WINDOW BETWEEN SUCCESSIVE ACTIONS DIVERSE FUNCTIONS DIVERSE PERSONNEL AND LOCATIONS References:  References IAEA-TECDOC-648 Procedures for conducting common cause failure analysis in probabilistic safety assessment (1992)

Add a comment

Related presentations

Related pages

Basic Level 1. PSA course for analysts - Nuclear Safety ...

Basic Level 1. PSA course for analysts Dependent failure analysis IAEA Training in level 1 PSA and PSA applications. Dependent failure analysis
Read more

Chapter 2 Temperature Dependent Ductile Material Failure ...

2 Temperature Dependent Ductile Material Failure Constitutive Modeling with Validation Experiments 11. ... 1. Wellman GW. Modeling ductile failure, ...
Read more

Modeling of Age-Dependent Failure Tendency from Incomplete ...

This paper addresses modeling of age-dependent failure rates from incomplete data that includes interval-censored failure ages. ... 1,2 Mio. Dokumente, ...
Read more

Modeling of Age-Dependent Failure Tendency from Incomplete ...

40.4 Modeling of Age-Dependent Failure Tendency. ... (1) The failure rate Λ′ is maximal at the age z = ( {{(1 + {1 mathord{left/ {vphantom ...
Read more

Chapter 40 Modeling of Age-Dependent Failure Tendency from ...

Modeling of Age-Dependent Failure Tendency from Incomplete Data ... Abstract This paper addresses modeling of age-dependent failure ... Table 40.1 The ...
Read more

Micromechanical Modeling of Time-Dependent Transverse ...

Micromechanical Modeling of Time-Dependent Transverse ... The time-dependent failure behaviour of transversely ... Mech. Time-Dependent Mater. 1, 1998 ...
Read more

Poly(ADP-ribose) catabolism triggers AMP-dependent ...

... the molecular mechanisms contributing to PARP-1-dependent impairment of ... prompting mitochondrial energy failure. ... Modeling and Simulation; Simcyp ...
Read more

Using the Fault Tree Method to Analyze Dependent and ...

Using the Fault Tree Method to Analyze Dependent and Independent Failure ... for modeling dependent and independent ... after 1 year (8760 hrs). The ...
Read more

Reliability Model of Mechanical Components with Dependent ...

A double integration model for reliability of mechanical components with dependent failure modes is ... dependent failure modes proposed is correct. 1.
Read more

Modelling the failure of reinforced concrete with nonlocal ...

1. Introduction. The failure process ... for the modelling of shear and compressive failure ... over an element length dependent ...
Read more