Component Unavailability Research Articles

Risk-informed design of advanced nuclear reactor: A target unavailability approach

Incorporating specified safety and production targets during the design phase of an engineered facility can effectively support reductions in construction and operation costs of nuclear power plants. In addition, this supports efforts toward making nuclear power technology more competitive and profitable than other electric power production forms by incorporating operational and design features into the design phase that directly support the increased likelihood of meeting such originating targets. Unavailability is the key parameter to establish plant-level capabilities that cascade down to function, system, and component levels to ensure that all sources of unavailability are addressed in the facility's design phases. In this article, two methods are presented to propose unavailability targets for nuclear reactor systems to risk-inform and optimize the design features, from an availability perspective: the first method is a simple method, and the second is a non-dominated sorting genetic algorithm III (NSGA-III) multi-objective optimization. These iterative methods are demonstrated in this paper for a hypothetical facility with a design maturity such that there is knowing the main components to comply with the engineering and security functions of the systems. With an assumed mature design, it is implied that generic data to execute quantitative reliability, and availability analyses are available at the systems levels that are sufficient to support a design-specific probabilistic risk assessment (PRA). The proposed methods modify the probabilities of basic events within the fault tree structures that model system and component unavailability sources that will be required for the facility’s operating license. The methods provided are applied to visualize the characteristics of the problem and the quantitative results. This work forms the basis for future work in which design alternatives affecting unavailabilities can be estimated and compared to their associated unavailability targets at different levels of granularity.

Effect of test-caused degradation on the unavailability of standby safety components

This paper proposes a safety-critical standby component unavailability model that contains aging effects caused by the elapsed time from installation, component degradation due to surveillance tests, and imperfect maintenance actions. An application of the model to a Motor-Operated Valve and a Motor-Driven Pump involved in the HPIS of a VVER/1000-V446 nuclear power plant is demonstrated and compared with other existing models at component and system levels. In addition, the effects of different unavailability models are reflected in the NPP's risk criterion, i.e., core damage frequency, over five maintenance periods. The results show that, compared with other models that do not simultaneously consider the full effects of degradation and maintenance impacts, the proposed model realistically evaluates the unavailabilities of the safety-related components and the involved systems as a plant age function. Therefore, it can effectively reflect the age-dependent CDF impact of a given testing and maintenance policy in a specified time horizon.

Component unavailability model for PSA Level I applications considering aging and startup degradation: An Application to an Emergency Diesel Generator System

The paper proposes an unavailability model, which considers standby component degradation due to repeated startups, including not only planned surveillance tests, but also additional startups due to real demands and, in cases when the component is detected failed in the planned test, extra tests related to corrective maintenance. The model is applied to evaluate the effect of additional tests on Emergency Diesel Generators from a Nuclear Power Plant’s Onsite Power System. These additional tests are not considered in traditional reliability models. The results show that excessive Diesel Generator testing during its working life may lead to a significant increase of component unavailability with reflections on the core damage frequency at the end of its overhaul maintenance period. This increment could justify anticipation of Diesel Generator overhaul maintenance in one or two working cycles.

Research on Time-Dependent Component Importance Measures Considering State Duration and Common Cause Failure

Unlike the current risk monitors, Real-time Online Risk Monitoring and Management Technology is characterized by time-dependent modeling on the state duration of components. Given the real-time plant configuration, it eventually provides the time-dependent risk level and importance measures for operation and maintenance management. This paper focuses on the assessment method of time-dependent importance measures and its risk-informed applications in real-time online risk monitoring and management technology, including Fussell-Vesely (FV), risk achievement worth (RAW), and risk reduction worth (RRW). In this study, the values of component importance have been investigated with a time-dependent risk quantification model, as well as the common cause failure treatment model. Here three options of common cause failure treatment have been developed, assuming that the unavailability of a component could be due to an independent factor (Option 1), a common cause factor (Option 2), or an unconfirmed cause (Option 3). In the special case of “what if a component is out-of-service” of the RAW numerator, a hybrid method for the RAW evaluation is presented resulting in a balanced and reasonable RAW value. A simple case study was demonstrated. The results showed that the absolute values and ranking order of time-dependent importance not only reflected the effect of the cumulative state duration of component on risk, but also comprehensively accounted for all possible situations of component unavailability. Moreover, time-dependent importance measures improved and provided novel insights for online configuration management, 1) ranking SSCs/events/human actions for controlling increased risk and optimizing near–term plans; and 2) exempting or limiting temporary configurations during online operation.

Time-dependent unavailability of equipment in an ageing NPP: Sensitivity study of a developed model

A previously developed model for assessing time-dependent unavailability of ageing safety equipment is briefly presented at the beginning of this paper. One of the essential features of this model is that it simultaneously considers the effects of ageing, testing, preventive and corrective maintenance and overhaul.The main focus of this paper is aimed towards performing sensitivity analysis of the developed model. A component level resolution is selected as the basis for performing the analysis. Investigation of the influence of different component-relevant input parameters on the calculated equipment unavailability is the goal of the analysis. The dependency of the calculated component unavailability on the corresponding surveillance test interval is of a particular interest. The focus is being given to one of the specifics of the developed model – the aggregation approach, i.e. the aggregation limiter that copes with the breach of computational memory.The results show that a relatively low value of the aggregation limiter implicates discrepancies in the shape of the component unavailability as a function of the surveillance test interval in some cases. These discrepancies are being substantially reduced with the increase of the value of the limiter. Consequently, longer computational time and higher memory requirements are directed.

Impact of component unavailability uncertainty on safety systems unavailability

The increased and extended application of the probabilistic safety assessment requires appropriate consideration of uncertainties in the analyses and interpretation of the results. Inadequate treatment of uncertainties may lead to poorly supported or even wrong conclusions with final consequence of loss of adequate level of safety.Epistemic uncertainty results from the imperfect knowledge or incomplete information regarding the parameters values in the underlying model. Epistemic uncertainty is considered in the probabilistic safety assessment models by probability distributions associated with the uncertain parameters.This paper presents the results of the analysis of the introduction of probability distributions associated with component unavailability parameters, on the overall unavailability of the analyzed system. The normal and lognormal distributions are introduced as probability distributions associated with component unavailability. The minimal cut sets of the analyzed system are identified with application of the fault tree analysis. The distribution of the top event unavailability is assessed with application of the min cut upper bound approximation and Monte Carlo sampling.The analysis of the uncertainty propagation is demonstrated using the fault tree of the auxiliary feedwater system of the nuclear power plant. The implications of the introduction of different probability distributions for the components unavailability on the obtained results are analyzed. Obtained results include skewness and kurtosis as measures of the goodness-of-fit of the top event unavailability to the normal distribution.Obtained results show that the probability density function of the top event depends on the characteristics of the basic events unavailability distribution and the importance of the selected events. Introduction of the lognormal distribution for uncertainty characterization of the basic events unavailability results in positive kurtosis. Positive kurtosis is indicating increased probability of having top event unavailability larger than the mean value.Decision making based on the mean value of the top event unavailability in case of positive kurtosis of the probability density function results in risk underestimation.

Integrated risk assessment for LNG terminals

This paper presents an integrated risk assessment framework for Liquefied Natural Gas (LNG) terminals. The basic steps for risk assessment are the following: a) hazard identification, b) accident sequence modeling, where logic models such as Event Trees and Fault Trees are developed c) data acquisition and parameter estimation, used to estimate frequencies of the initiating events, component unavailability and probabilities of human actions, d) accident sequence quantification, where all accident sequences are assessed, e) consequence assessment, where release, evaporation rate, radiation levels and overpressure owing to immediate or delayed ignition of LNG is performed and f) integration of results where risk indices are assessed. Risk assessment of an onshore and an offshore LNG terminal is performed, according to the basic methodological steps, and the distances where individual risk levels equal to 10−5–10−7/yr are assessed and presented in the form of individual contours. A section dedicated to I.A. Papazoglou, pioneer in Quantitative Risk Assessment both in the nuclear and chemical industry is also provided.

Integration of Degradation Models Into Generation Risk Assessment: Challenges and Modeling Approaches

The main objective of generation risk assessment (GRA) is to assess the impact of equipment unavailability and failures on the ability of the plant to produce power over time. The system reliability models employed for this purpose are based on the standard fault tree/event tree approach, which assumes failure rates to be constant. However, this ignores the impact of aging degradation and results in static estimates of expected generation loss. Component and equipment degradation not only increases the probability of failure over time, but also contributes to generation risk through increased unavailability and costs arising from greater requirement for inspection and replacement of degraded components. This paper discusses some of the key challenges associated with integrating the results of component degradation models into GRA. Because many analytical and simulation methods are subject to limitations, the methodology and modeling approach proposed in this work builds on the current GRA practice using the fault tree approach. The modeling of component degradation can be done separately at the fault tree cut set level, assuming the cut sets are independent and the component unavailabilities are relatively small. In order to capture the joint contribution of equipment failure and unavailability to generation risk, new risk-based importance measures are also developed using the concept of net present value.

Analytical model for unavailability due to aging failures in power systems

Power system component failures can generally be classified into two categories: repairable random failures and aging failures. An essential step in power system reliability evaluation is the calculation of component unavailability. The paper presents a new model for calculating the unavailability due to aging failures using the normal distribution. It is based on the strictly mathematical derivation and can be used more easily with high accuracy compared to the traditional model. An example of seven generating units and two test systems of the RBTS and the IEEE-RTS were used to demonstrate effectiveness of the proposed model and its applications in power system reliability evaluation.

Etudes de sensibilité, facteurs d'importance et défaillances de cause commune

In this paper, we derive mathematical expressions to assess reliability importance measures for a component in a common cause failure group (CCF). The main originality of this work is that the existing information on the cause of the component unavailability (intrinsic, unknown, CCF-related, maintenance) can be explicitly taken into account in the importance factor computation.

A genetic algorithm solution for a nuclear power plant risk–cost maintenance model

Reliability Centered Maintenance (RCM) is one of the popular maintenance optimization methods according to certain kinds of priorities. Traditional RCM usually analyzes and optimizes the maintenance strategy from the viewpoint of component instead of the whole maintenance program impact. Research presented in this paper is a pilot study using PSA techniques in RCM. How to reflect the effect on component unavailability by the maintenance activities such as surveillance testing and preventive maintenance in PSA model is discussed firstly. Based on the discussion, a maintenance risk–cost model is established for global maintenance optimization in a nuclear power plant, and a genetic algorithm (GA) is applied to solve such a model to get the global optimized maintenance strategy. Finally, the result got from a simple test case based on a risk–cost model consisting of 10 components is presented.

Treatment of general dependencies in system fault-tree and risk analysis

Implicit and explicit methods are described for reliability and risk analysis of systems with dependent or correlated basic events. General rules are presented for modeling any group of n mutually s-dependent events with 2/sup n/-1 s-independent events. The of these virtual events are determined based on the joint of the original s-dependent events, typically known by s-correlation or conditional probabilities. The transformations preserve the values of all terms (e.g., minimal cut sets), independent of system success criteria. This facilitates general use of ordinary fault-tree computer codes that assume basic events to be s-independent. Explicit basic event are obtained for calculating the probability of failure on demand of standby safety systems when the s-dependency is caused by scheduling and synchronization of test episodes between n redundant components (1 /spl les/ n /spl les/ 4), and by statistical variation of failure rates. Interesting probabilities are encountered in this exercise, mainly due to negative s-correlation between the component unavailabilities with staggered testing. Results obtained for human-error events are useful when the conditional probability to repeat an error is larger than the probability of an error in a single isolated task. Explicit results are obtained for systems with time-related common-cause failures modeled by general multiple failure rates. The impacts of test intervals and test staggering are included. Staggered testing is optimal with an ETR (extra-testing rule), although ETR is not important for 1-out-of-n:G systems. An economic model provides insights into the impacts of various parameters: the optimal test interval increases with increasing redundancy and testing cost, and it decreases with increasing accident cost and initiating event rate. Staggered testing with ETR allows for the longest optimal test intervals. Rules are presented for changing s-dependency when some component is known to be failed. Current fault-tree quantification tools are not well geared to use the implicit method in spite of the fact that it would simplify the fault-tree construction, reduce the number of cut sets, and allow different types of dependencies or correlations in the analysis. A recommendation is to computerize the implicit method or include it as an option to current codes. It would need only a data table for joint and the ability to pick-up data from this table whenever two or more of the s-dependent events appear in a term (or a cut set).

An application of genetic algorithms to surveillance test optimization of a PWR auxiliary feedwater system

Nuclear power plant systems are comprised of both on-line and standby components. Standby components differ from on-line ones, as they might be unavailable due to unrevealed failures. The usual procedure employed to reveal failures before real demands is to submit the component to surveillance tests. Surveillance test policies might deal with two conflicting scenarios: the test frequency must be sufficiently high in order to reveal failures before demands, but, on the other hand, it must be low enough due to its influence on the component unavailability. Standard surveillance test policies for typical nuclear power plants usually consist of periodic tests for which the frequencies are often higher than necessary for obtaining the optimal availability. In this work, a new surveillance test policy optimization method, based on genetic algorithms, is applied to the Angra-I (Brazilian PWR) auxiliary feedwater system. The new probabilistic model has been developed in order to comprise the following features: (1) aging effects on standby components when they undergo surveillance tests; (2) revealing failures during the surveillance tests implies corrective maintenance, and, consequently, increasing outage times; (3) components are distinct (i.e., each has distinct test parameters, such as outage time, aging factors, etc); (4) tests are not necessarily periodic. The results, when compared to those obtained by standard test policies, show improved overall availability at the system level. © 2002 Wiley Periodicals, Inc.

Aging Effects on Time-Dependent Nuclear Plant Component Unavailability: An Investigation of Variations from Static Calculations

In the nuclear industry, aging effects have been traditionally incorporated into probabilistic risk assessment studies by using a constant (static) unavailability (q s ) averaged over time. However, recent work shows that because of aging, substantial deviations may occur in time-dependent nuclear plant component unavailability from that predicted by static models well within the plant lifetime. A methodology based on the standard extension of the classic renewal equation when repair is explicitly considered is used to investigate (a) the trends in the effects of aging on time-dependent component unavailability as a function of changing first failure density (FFD) and test parameters and (b) the circumstances for which static approximations may be inadequate to describe these effects. The investigation uses several point- and time-averaged unavailability measures based on time-dependent unavailability, such as before-test unavailability (BTU), average-interval unavailability (AIU) and year-average unavailability (YA U), and is restricted to periodically tested components whose FFDs satisfy the Weibull distribution with aging threshold. The results show that while point measures (e.g., BTU) can substantially differ from static unavailability and while all measures are sensitive to changes in the Weibull shape parameter b, aging threshold time τ, and time between tests T, the differences between the time-averaged measures used (e.g., AIU, YA U) and the static unavailability were only found to be relatively significant for one case among more than 100 combinations of b, τ, and T that were investigated. The differences area factor of 18 months) and may describe the late effects ofaging on component unavailability irrespective of b and T (i. e., beyond 25 yr of component age for the data under consideration).

Modelling common cause failures under data uncertainty of the system unavailability

This paper uses Yeh's work to model common cause failures (CCFs) during the data uncertainty analysis of the system unavailability. The unavailabilities due to independent failure of the equipments sharing a common environment are regarded during the uncertainty analysis as correlated variables in modelling common cause failures due to a known stochastic coupling. The correlation between the unavailability of the system due to independent failure and that due to definite and simultaneous unavailability of all the equipments of the system because of a common cause, has also been considered. Further, a method has been described to simulate the values of correlated log-normal variables. The results of the simulation studies carried out using this method for a 2-out-of-3 system have also been described. It was found, as expected, that as correlation co-efficients between unavailabilities of component unavailabilities or between those due to independent and dependent failures increase, the dispersion of the distribution of the system unavailability increases. Also, the distribution tends to become log-normal. It has been pointed out that, while several other approaches for common cause analysis can provide “point estimates”, the approach studied here helps in the treatment of data uncertainties. The approach of this paper appears to have a sounder physical and statistical basis than most of the existing models and is simpler as it requires only two additional parameters to be estimated for a redundant system with any number of identical components, while other models require more.

Analysis of Uncertainties in Interactive Probabilistic Safety Analysis (PSA) Models

Probabilistic safety analysis (PSA) has been used in nuclear, chemical, petrochemical, and several other industries. The probability and/or frequency results of most PSAs are based on average component unavailabilities during the mission of interest. While these average results are useful, they provide no indication of the significance of the facility's current status when one or more components are known to be out of service. Recently, several interactive computational models have been developed for nuclear power plants to allow the user to specify the plant's status at a particular time (i.e., to specify equipment known to be out of service) and then to receive updated PSA information. As with conventional PSA results, there are uncertainties associated with the numerical updated results. These uncertainties stem from a number of sources, including parameter uncertainty (uncertainty in equipment failure rates and human error probabilities). This paper presents an analysis of the impact of parameter uncertainty on updated PSA results.

Quantifying maintenance effects on unavailability and risk using Markov modeling

Abstract A Markov approach is presented for quantifying the effects of maintenance on unavailability and risk. Markov modeling is standard; however, what is new is the new applications that are presented, including the relations which are developed for the required transition rates to allow practical implementations of the model. Maintenance effects are quantified by defining a degraded state for the component in addition to an operational state and a failed state. The Markov maintenance model which is developed is a natural extension of the standard models used in Probabilistic Risk Assessments (PRAs) and simplifies to the PRA models when the degraded state is not differentiated from the operational state. Identification of a degraded state allows the benefits of maintenance to be explicitly evaluated and allows optimal maintenance intervals to be determined. The component unavailabilities which are determined can be subsequently used in a PRA to evaluate the risk effectiveness of maintenance. Applications of the model are demonstrated. The demonstrations indicate that maintenance effects on component unavailability can be significant in certain situations.

Finding the exact top event unavailability mean and variance expressions of fault trees having correlated jointly lognormal component failure probabilities

Abstract This paper describes related extensions to two methods. One method is given by Apostolakis and Lee 1 and uses minimal cut-sets and Taylor Series expansions to estimate the mean and variance of the top-event unavailability (probability of being unavailable) of fault trees. The other method is given by Zhang 2 for estimating the mean and variance with Monte Carlo when the component unavailability distributions are correlated and the correlations are decomposable. The extensions produce analysable expressions that handle correlated component unavailability probabilities. The expressions are exact when, as is often done, 1,3 the component unavailability probabilities are modeled using jointly lognormal distributions. For Zhang, an added condition for exactness is that the decomposition is in terms of products of exponentiated underlying lognormal random variables.

The human element

In past issues we have discussed various system-related disasters and their causes, both accidental and intentional. In almost all cases it is possible to allocate to people—directly or indirectly—those difficulties allegedly attributed to “computer problems.” But too much effort seems directed at placing blame and identifying scapegoats, and not enough on learning from experiences and avoiding such problems [1,2,5,6,7]. Besides, the real causes may implicitly or explicitly involve a multiplicity of developers, customers, users, operators, administrators, others involve with computer and communication systems, and sometimes even unsuspecting bystanders. In a few cases the physical environment also contributes, e.g., power outages, floods, extreme weather, lightning, and earthquakes. Even in those cases there may have been system people who failed to anticipate the possible effects. In principle, at least, we can design redundantly distributed systems that are able to withstand certain hardware faults, component unavailabilities, extreme delays, human errors, malicious misuse, and even “acts of God”—at least within limits. Nevertheless, in surprisingly many systems (including systems designed to provide continuous availability), an entire system can be brought to a screeching halt by a simple event just as by a complex one [4].

PRAAGE88: An interactive IBM-PC code for aging analysis of NUREG-1150 systems

The effects of time dependent failure rates caused by the aging of components are becoming increasingly important in probabilistic risk assessment (PRA) and reliability analyses of nuclear power plant systems. In the NRC Nuclear Plant Aging Research (NPAR) program, the effects of aging in nuclear systems are being evaluated through the use of time varying failure rates that are determined as a function of the age of the system. These analyses involve complex systems and include various sensitivity studies; thus, the PRAAGE88 computer code was developed to facilitate these calculations. PRAAGE88 is an IBM PC based code that computes system unavailability, component unavailability, and various importance measures for use in evaluating the effect of aging on reactor systems. This paper describes the methodology utilized in the code, its capabilities and areas of application.