Probabilistic Risk Assessment Framework Research Articles

Dynamic probabilistic risk assessment of decision-making in emergencies for complex systems, case study: Dynamic positioning drilling unit

Decision-making in emergency situations is a risky and uncertain process due to the limited information and lack of time. Some key problem parameters, such as the time required to complete important response tasks, must be estimated and are therefore prone to errors. Other parameters, such as the probability of occurrence of a consequential event, will typically change as the response operation progresses. As a result, there should be a dynamic probabilistic risk assessment framework to assess the risk level of decision scenarios and facilitate the decision-making process.In this paper, a methodology for dynamic probabilistic risk assessment of decision making in emergencies for complex marine systems is proposed. In this method, a dynamic event sequence diagram is introduced that helps to quantify events probabilities as a function of time, as well as environmental and operational variables, considering events interdependencies and uncertainties. In addition, the effects of time required11Time required refers to the time needed to perform each level of decision-making process (detection, diagnosis, decision-making, execution). and time available22Time available is the time remaining before an accident happens in a system. It is derived from the dynamic simulator of the system. The dynamic simulator calculates the time remaining before a collision based on the system components status, operation and environmental conditions at the time of incident. for performing a decision in emergency are considered in the risk model. In this methodology, probabilistic models including Bayesian network and Monte Carlo simulation are utilized to quantify the uncertain behavior of the decision-making process in complex marine systems.A computational study is also conducted to evaluate the methodology performance, in terms of effectiveness and efficiency. Computational results show that the proposed approach can obtain optimal solutions for large and practical problem sizes.

Risk Assessment of Future Climate and Land Use/Land Cover Change Impacts on Water Resources

Climate and land use and land cover (LULC) changes will impact watershed-scale water resources. These systemic alterations will have interacting influences on water availability. A probabilistic risk assessment (PRA) framework for water resource impact analysis from future systemic change is described and implemented to examine combined climate and LULC change impacts from 2011–2100 for a study site in west-central Texas. Internally, the PRA framework provides probabilistic simulation of reference and future conditions using weather generator and water balance models in series—one weather generator and water balance model for reference and one of each for future conditions. To quantify future conditions uncertainty, framework results are the magnitude of change in water availability, from the comparison of simulated reference and future conditions, and likelihoods for each change. Inherent advantages of the framework formulation for analyzing future risk are the explicit incorporation of reference conditions to avoid additional scenario-based analysis of reference conditions and climate change emissions scenarios. In the case study application, an increase in impervious area from economic development is the LULC change; it generates a 1.1 times increase in average water availability, relative to future climate trends, from increased runoff and decreased transpiration.

Reverse the declining course: A risk assessment for marine and fisheries policy strategies in Europe from current knowledge synthesis

An ecosystem approach to fisheries management is being developed around the world in an attempt to consider components of marine ecosystems other than just the exploited stocks. While numerous scientific studies on the consequences of fishing on marine ecosystems exist, most of their findings are too uncertain and fail to quantify the magnitude of the effects they describe to be used reliably in an environmental management and marine policy-making realm. To circumvent such a knowledge gap, we built and fitted a Bayesian network (BN), informed by a review of past studies, which integrates the links and direction of effects between socio-ecosystem components. These effects are represented as conditional probabilities, so that missing magnitudes of some ecosystem effects emerge from the numerous past cases that were reviewed. The Bayesian network is informed by a collection of cases extracted from 246 published scientific studies investigating relationships in marine ecosystems. We find that marine ecosystems are likely to be on a declining course under conjugated pressures of both fishing and changes to environmental conditions, e.g. due to ongoing climate change. By querying the fitted BN to obtain posterior probabilities under different scenarios, we showed that increasing fisheries regulation and environmental governance could partly mitigate these effects and decrease the risk of biodiversity loss, decreased profit and social inequity; the three pillars of the EU Common Fishery Policy (CFP). We discuss these findings with regard to particular fisheries across EU Waters, specifically: the North Sea, the Baltic Sea, the North Western and the South Western Waters. Furthermore, we discuss how fishing impacts interact with other ecosystem effects and pressures, caused by, or causing possible far-reaching consequences in marine ecosystem dynamics. The Bayesian network has some limitations regarding the ability to handle feedback loops that can occur in natural marine ecosystems. Nevertheless, such an approach helps to take a holistic view and integrate existing knowledge and new findings from future work, in a coherent, probabilistic risk assessment framework, while identifying what leverage and management actions may help nudge the system toward desired states.

Marine mussel-based biomarkers as risk indicators to assess oceanic region-specific microplastics impact potential

Microplastics (MPs) pollution in the ocean is an area of growing concern. Marine fauna subject to MPs exposure from various environmental sources are most likely to have detrimental effects on their immune system. However, studies of potential risks of MPs on marine ecosystems in light of environmental concentrations are largely limited. To this end, we presented an approach for assessing potential impact of MPs on marine ecosystems based on marine mussel Mytilus-based probabilistic risk assessment framework. The immunotoxic-based biomarkers of Mytilus were used to assess the impact of MPs on marine environment appraised with datasets by compiling oceanic region-specific and comprehensive MPs−environment studies along with published toxicity experiments. The immunological effects of MPs on lysosomal destabilization and phagocytosis in hemocytes of Mytilus were reconstructed as the concentration–response functions. We assessed the risk for marine environment exceeding a threshold of Mytilus-based immunological toxicity based on the benchmark concentration (BMC) approach corresponding to a 10% inhibition effect (BMC10). We estimated BMC10 values for inhibitions of lysosomal membrane stability and phagocytosis to be 1.4 and 0.4 mg L−1, respectively. Here we showed that the overall MPs-associated impact potential was low among South Pacific Ocean, Mediterranean Sea, and South Atlantic Ocean. However, we found that MPs from North Pacific Ocean were very likely (>90% probability) to pose a potential threat to marine mussels. Our findings have important implications for understanding the linked relationships between environmental MPs and likelihood of exposure risk for marine organisms in different oceanic regions around the world. We suggest that Mytilus-based risk indicator for estimating risk metrics of essential marine ecosystems posed by environmentally relevant MPs concentrations can help inform practices for the sustainable management and for mitigating the environmental MPs-induced negative impact on marine organisms.

A probabilistic risk assessment framework considering lane-changing behavior interaction

Understanding the dynamic characteristics of surrounding vehicles and estimating the potential risk of mixed traffic can help reliable autonomous driving. However, the existing risk assessment methods are challenging to detect dangerous situations in advance and tackle the uncertainty of mixed traffic. In this paper, we propose a probabilistic driving risk assessment framework based on intention identification and risk assessment of surrounding vehicles. Firstly, we set up an intention identification model (IIM) via long short-term memory (LSTM) networks to identify the intention possibility of the surrounding vehicles. Then a risk assessment model (RAM) based on the driving safety field is employed to output the potential risk. Specifically, driving safety field can reflect the coupling relationship of drivers, vehicles, and roads by analyzing their interaction. Finally, an integrated risk evaluation model combining both IIM and RAM is developed to form a dynamic potential risk map considering multi-vehicle interaction. For example, in a typical but challenging lane-changing scenario, an intelligent vehicle can assess its driving status by calculating a risk map in real time that represents the risk generated by the estimated intentions of surrounding vehicles. Furthermore, simulations and naturalistic driving experiments are conducted in the extracted lane-changing scenarios, and the results verify the effectiveness of the proposed model considering lane-changing behavior interaction.

Site-specific machine learning predictive fertilization models for potato crops in Eastern Canada.

Statistical modeling is commonly used to relate the performance of potato (Solanum tuberosum L.) to fertilizer requirements. Prescribing optimal nutrient doses is challenging because of the involvement of many variables including weather, soils, land management, genotypes, and severity of pests and diseases. Where sufficient data are available, machine learning algorithms can be used to predict crop performance. The objective of this study was to determine an optimal model predicting nitrogen, phosphorus and potassium requirements for high tuber yield and quality (size and specific gravity) as impacted by weather, soils and land management variables. We exploited a data set of 273 field experiments conducted from 1979 to 2017 in Quebec (Canada). We developed, evaluated and compared predictions from a hierarchical Mitscherlich model, k-nearest neighbors, random forest, neural networks and Gaussian processes. Machine learning models returned R2 values of 0.49-0.59 for tuber marketable yield prediction, which were higher than the Mitscherlich model R2 (0.37). The models were more likely to predict medium-size tubers (R2 = 0.60-0.69) and tuber specific gravity (R2 = 0.58-0.67) than large-size tubers (R2 = 0.55-0.64) and marketable yield. Response surfaces from the Mitscherlich model, neural networks and Gaussian processes returned smooth responses that agreed more with actual evidence than discontinuous curves derived from k-nearest neighbors and random forest models. When conditioned to obtain optimal dosages from dose-response surfaces given constant weather, soil and land management conditions, some disagreements occurred between models. Due to their built-in ability to develop recommendations within a probabilistic risk-assessment framework, Gaussian processes stood out as the most promising algorithm to support decisions that minimize economic or agronomic risks.

Natural springs protection and probabilistic risk assessment under uncertain conditions

The study introduces a comprehensive framework for natural springs' protection and probabilistic risk assessment in the presence of uncertainty associated with the characterization of the groundwater system. The methodology is applied to a regional-scale hydrogeological setting, located in Northern Italy and characterized by the presence of high-quality natural springs forming a unique system whose preservation is of critical importance for the region. Diverse risk pathways are presented to constitute a fault tree model enabling identification of all basic events, each associated with uncertainty and contributing to an undesired system failure. The latter is quantified in terms of hydraulic head falling below a given threshold value for at least one amongst all active springs. The workflow explicitly includes the impact of model parameter uncertainty on the evaluation of the overall probability of system failure due to alternative groundwater extraction strategies. To cope with conceptual model uncertainty, two models based on different reconstructions of the aquifer geological structure are considered. In each conceptual model, hydraulic conductivities related to the geomaterials composing the aquifer are affected by uncertainty. It is found that (a) the type of conceptual model employed to characterize the aquifer structure strongly affects the probability of system failure and (b) uncertainties associated with the ensuing conductivity fields, even as constrained through model calibration, lead to different impacts on the variability of hydraulic head levels at the springs depending on the conceptual model adopted. The results of the study demonstrate that the proposed approach enables one to (i) quantify the risk associated with springs depletion due to alternative strategies of aquifer exploitation; (ii) quantify the way diverse sources of uncertainty (i.e., model and parameter uncertainty) affect the probability of system failure; (iii) determine the optimal exploitation strategy ensuring system functioning; and (iv) identify the most vulnerable springs, where depletion first occurs.

Wind and rain losses for metal-roofed contemporary houses subjected to non-cyclonic windstorms

Severe windstorms cause millions in losses annually for housing in Southeast Australia that has more than half of Australia’s population. The risk assessment for housing in these non-cyclonic regions is the key to assessing the cost-effectiveness of relevant wind mitigation measures to reduce economic losses. This study develops a probabilistic risk assessment framework to evaluate the wind and rain losses for new Australian contemporary houses correctly built and inspected to current standards that are subjected to non-cyclonic windstorms, which integrates the hazard modelling for extreme wind and associated rainfall, reliability-based wind damage assessment, rainwater intrusion evaluation and economic loss modelling. The risk analysis was conducted for metal-roofed contemporary houses in Brisbane and Melbourne, and the efficacy of the proposed risk assessment framework was demonstrated by comparing with inferred insurance loss data. It was found that rainwater damage to building interior and contents is the major contributor to annual expected economic losses associated with windstorms, whereas wind damage to roof cladding and windows comprises a small portion of annual losses. Preliminary model outputs also indicate that houses in Brisbane are generally subject to more losses than houses in Melbourne. However, modelling assumptions that lead to these results have yet been fully validated.

Probabilistic human health risk assessment of perfluorooctane sulfonate (PFOS) by integrating in vitro, in vivo toxicity, and human epidemiological studies using a Bayesian-based dose-response assessment coupled with physiologically based pharmacokinetic (PBPK) modeling approach

BackgroundEnvironmental exposure to perfluorooctane sulfonate (PFOS) is associated with various adverse outcomes in humans. However, risk assessment for PFOS with the traditional risk estimation method is faced with multiple challenges because there are high variabilities and uncertainties in its toxicokinetics and toxicity between species and among different types of studies. ObjectivesThis study aimed to develop a robust probabilistic risk assessment framework accounting for interspecies and inter-experiment variabilities and uncertainties to derive the human equivalent dose (HED) and reference dose for PFOS. MethodsA Bayesian dose-response model was developed to analyze selected 34 critical studies, including human epidemiological, animal in vivo, and ToxCast in vitro toxicity datasets. The dose-response results were incorporated into a multi-species physiologically based pharmacokinetic (PBPK) model to reduce the toxicokinetic/toxicodynamic variabilities. In addition, a population-based probabilistic risk assessment of PFOS was performed for Asian, Australian, European, and North American populations, respectively, based on reported environmental exposure levels. ResultsThe 5th percentile of HEDs derived from selected studies was estimated to be 21.5 (95% CI: 10.6–36.3) ng/kg/day. After exposure to environmental levels of PFOS, around 50% of the population in all studied populations would likely have >20% of increase in serum cholesterol, but the effects on other endpoints were estimated to be minimal (<10% changes). There was a small population (~10% of the population) that was highly sensitive to endocrine disruption and cellular response by environmental PFOS exposure. ConclusionOur results provide insights into a complete risk characterization of PFOS and may help regulatory agencies in the reevaluation of PFOS risk. Our new probabilistic approach can conduct dose-response analysis of different types of toxicity studies simultaneously and this method could be used to improve risk assessment for other perfluoroalkyl substances (PFAS).

Human Reliability Analysis-Based Method for Manual Fire Suppression Analysis in an Integrated Probabilistic Risk Assessment

Abstract Fire is one of the most critical initiating events that can lead to core damage in nuclear power plants (NPPs). To evaluate the potential vulnerability of plants to fire hazards, fire probabilistic risk assessment (PRA) is commonly conducted. Manual fire protection features, performed by the first responders (e.g., fire brigade), play a key role in preventing and mitigating fire-induced damage to the plant systems. In the current fire PRA methodology of NPPs, there are two main gaps in the modeling of manual fire protection features: (i) the quantification of the first responder performance is solely based on empirical data (industry-wide historical fire events), and so the plant-specific design and conditions cannot be explicitly considered; and (ii) interactions of first responders with fire propagation are not fully captured. To address these challenges, the authors develop a model-based approach, grounded on human reliability analysis (HRA) and coupled with the fire dynamics simulator (FDS), to model the first responder performance more realistically and consider the interface between the first responder performance and fire propagation more explicitly. In this paper, the HRA-based approach is implemented in an integrated PRA (I-PRA) methodological framework for fire PRA and applied to a switchgear room fire scenario of an NPP. The proposed model-based approach (a) adds more realism to fire PRA and so to risk assessment in NPPs and (b) provides opportunities for sensitivity and importance measure analyses with respect to design conditions; therefore, contributes to risk management in NPPs.

Global importance measure methodology for integrated probabilistic risk assessment

An integrated probabilistic risk assessment framework combines spatio-temporal probabilistic simulations of underlying failure mechanisms with classical probabilistic risk assessment logic consisting of event trees and fault trees. The importance measure methods commonly used in classical probabilistic risk assessment, for example, Fussell–Vesely importance measure and Risk Achievement Worth, generate the ranking of components at the basic event level utilizing the one-at-a-time and local methods. These classical importance measure methods, however, are not sufficient for integrated probabilistic risk assessment where, in addition to the component-level risk importance ranking, the ranking of risk-contributing factors (e.g. physical design parameters) associated with the underlying failure mechanisms is desired. In this research, the global importance measure method is suggested and implemented for the integrated probabilistic risk assessment framework. The global importance measure can account for four key aspects of integrated probabilistic risk assessment, including (a) ranking of input parameters at the failure mechanism level based on the contribution to the system risk metrics, (b) uncertainty of input parameters, (c) non-linearity and interactions among input parameters inside the model, and (d) uncertainty associated with the system risk estimates; therefore, it enhances the accuracy of risk importance ranking when the risk model has a high level of non-linearity, interactions, and uncertainties. This article shows (a) qualitative justifications for the selection of the global importance measure method for the integrated probabilistic risk assessment framework and (b) quantitative proof of concept using three case studies, including two illustrative fault tree examples and one practical application of the integrated probabilistic risk assessment framework for Generic Safety Issue 191 at nuclear power plants (a sump blockage issue following a loss-of-coolant accident).

Measuring risk-importance in a Dynamic PRA framework

Risk Importance Measures (RIMs) are indexes that are used to rank Structures, Systems, and Components (SSCs). The most used measures are: Risk Reduction Worth, Risk Achievement Worth, Birnbaum and Fussell-Vesely. Once obtained from Classical Probabilistic Risk Assessment (PRA), these risk measures can be effectively employed to identify the most risk-important SSCs. The objective of this paper is to present a series of methods that can be employed to measure risk importance of SSCs from Dynamic PRA. In contrast to Classical PRA methods, Dynamic PRA methods couple stochastic models with system simulators to determine risk associated to complex systems such as nuclear plants. Compared to Classical PRA methods, Dynamic PRA approaches can evaluate with higher resolution the safety impact of timing and sequencing of events on the accident progression. The developed set of RIMs are directly derived from Classical RIMs and adapted to deal with simulation-based data. We present a series of analytical tests to show how RIMs can be obtained from a Dynamic PRA data and a comparison of the RIMs obtained from Classical and Dynamic PRA for a Large Break Loss Of Coolant Accident (LB-LOCA) initiating event of a Pressurized Water Reactor (PWR). The obtained results have highlighted differences among the two PRA approaches in the predicted final outcome of few accident sequences. This have consequently affected the risk importance of a subset of basic events.

Risk assessment for a long-span cable-stayed bridge subjected to multiple support excitations

Current probabilistic seismic risk assessment framework does not fully address the impact of the spatial variability of ground motions on long bridges. However, recent studies have highlighted that long-span bridges are more vulnerable to damage due to spatial varying ground motions. To achieve this increased vulnerability, this research presents a reliable risk assessment tool for long-span bridges subjected to spatial varying ground motions. The proposed method includes the following tasks: (i) the simulation of spatially variable ground motions, (ii) creation of numerical bridge model, (iii) generation of probabilistic seismic demand models, (iv) development of component and system fragility curves, (v) estimation of annual exceedance of damage through site-compatible hazard curves, and (vi) computation of repair time. As an illustration, a cable-stayed bridge is selected to highlight the effect of spatial variability of ground motions on the seismic risk and repair time of individual bridge components and bridge system in comparison to uniform excitations. Additionally, a sensitivity analysis is performed to examine the effect of spatial variability parameters and hazard curve-related parameters on the seismic risk of the bridge.

Risk-based approach towards design evaluation and re-assessment of shutdown safety margin

Traditional approach for design, operation and regulation of the nuclear plants is deterministic in nature where the principle of defense in depth governs incorporation of multiple barriers and levels of protection. Single failure criteria, redundancy, diversity, fail safe criteria, quality assurance, etc., form the major cornerstone of this approach for safety demonstration. Traditional approach, which is conservative and prescriptive in nature, has served well for high level of safety since the inception of nuclear industry. However, with the accumulation of operating experiences, insights obtained from the accidents in the nuclear plants and growth of the probabilistic risk assessment methods, there is an increasing trend to implement a risk-informed approach in the nuclear plants. Over conservatism built in design, operation and regulation can be addressed using quantitative insights on risk and uncertainty such that the system availability or performance can be enhanced without compromising the safety. This paper proposes an integrated risk-based approach wherein probabilistic risk assessment framework has been used along with the deterministic insights to give credit to the available design provision in the nuclear plants. The objective is to use the margins available with the existing design to demonstrate safety of the plant by using risk framework along with the improved understanding of uncertainty. The model and procedure, involved in the risk-based approach, have been demonstrated through a case study performed on an Indian research reactor.

Beneficial and Detrimental Effects of Soil-Structure Interaction on Probabilistic Seismic Hazard and Risk of Nuclear Power Plant

The purpose of this study is to investigate the soil-structure interaction (SSI) effect on the overall risk of a PWR containment building structure with respect to two failure modes: strength and displacement. The precise quantification of the risk within the seismic probabilistic risk assessment framework depends considerably on an accurate treatment of the seismic response analysis. The SSI effect is one of the critical factors to consider when accurately predicting structural responses in the event of an earthquake. Previous studies have been conducted by focusing more on the positive side of the SSI effects and the effects mainly on the seismic fragility result. Therefore, this paper presents the results of a study of the SSI effect on the overall risk. Also, the study relies on an emphasis on revealing a beneficial and a detrimental effect of the SSI by utilizing an example of the containment structure in three soil conditions and two main failure modes. As a result, the consideration of SSI shows a complete conflicting effect on the seismic fragility and risk results depending on two failure modes considered in this study. This has a positive effect regarding the strength failure mode, but this brings a negative effect regarding the displacement failure mode. The risk fluctuation width is particularly noticeable in the site having a considerable change in seismic hazard information such as Los Angeles on the western site of the US. Such results can be expected to be utilized in a future study for investigating the pros and cons of the SSI effect associated with various failure modes in diverse conditions.

Analytical tsunami fragility curves for seaport RC buildings and steel light frame warehouses

Harbours are crucial assets for the sustainment and development of human activities. The recent devastating tsunami events as well as the increasing number of people, structures and economic activities being exposed to tsunami hazards revealed the need for the estimation of the effects of tsunami wave on seaport structures. However, only a limited number of tools to estimate the potential impacts of tsunami are available until now. This study aims at developing analytical tsunami fragility functions for some representative typologies of seaport structures in Greece. In particular, low-code moment resisting frame (MRF) and dual reinforced concrete (RC) buildings of various heights, and a typical warehouse are considered in the analysis. A numerical investigation is performed considering different combinations of tsunami loads based on FEMA P646 (2008) [1] recommendations for gradually increasing tsunami inundation depths and for the various structure typologies. To minimize the uncertainties related to the definition of damage limit states, tsunami nonlinear static analyses are performed and appropriate tsunami capacity curves are derived for the considered structures. Structural limit states are defined on tsunami capacity curves in terms of threshold values of material strain. For the complete damage state, shear failure is also considered, since the collapse of structures may be caused by the occurrence of either a flexural or shear failure in structural components. Fragility curves are numerically calculated for the different damage states using nonlinear regression analysis. They could be used within a probabilistic risk assessment framework to assess the vulnerability of low-code RC buildings and typical warehouses exposed to tsunami hazard along European-Mediterranean and other regions of similar facilities worldwide.

Mining data in a dynamic PRA framework

Computational, also known as Dynamic, Probabilistic Risk Assessment (PRA) methods employ system simulation codes coupled with stochastic analysis tools in order to determine probabilities of certain outcomes such as system failure. In contrast to Classical PRA methods (i.e., Event-Tree and Fault-Tree) in which timing and sequencing of events is set by the analyst, accident progression is dictated by the system control logic and its interaction with the system temporal evolution. Due to the nature of the problem, Dynamic PRA methods can be expensive form a computational point of view since a large number of accident scenarios is simulated. Consequently, they also generate a large amount of data (database storage may be on the order of gigabytes or higher). We investigate and apply several methods and algorithms to analyze these large time-dependent data sets. The objective is to present a broad overview of methods and algorithms that can be used to improve data quality and to analyze and extract information from large data sets containing time dependent data. In this context, “extracting information” means constructing input-output correlations, finding commonalities, and identifying outliers.

Probabilistic risk assessment based model validation method using Bayesian network

Past few decades have seen a rapid growth in the availability of computational power and that induces continually reducing cost of simulation. This rapidly changing scenario together with availability of high precision and large-scale experimental data has enabled development of high fidelity simulation tools capable of simulating multi-physics multi-scale phenomena. At the same time, there has been an increased emphasis on developing strategies for verification and validation of such high fidelity simulation tools. The problem is more pronounced in cases where it is not possible to collect experimental data or field measurements on a large-scale or full scale system performance. This is particularly true in case of systems such as nuclear power plants subjected to external hazards such as earthquakes or flooding. In such cases, engineers rely solely on simulation tools but struggle to establish the credibility of the system level simulations. In practice, validation approaches rely heavily on expert elicitation. There is an increasing need of a quantitative approach for validation of high fidelity simulations that is comprehensive, consistent, and effective. A validation approach should be able to consider uncertainties due to incomplete knowledge and randomness in the system's performance as well as in the characterization of external hazard. A new approach to validation is presented in this paper that uses a probabilistic index as a degree of validation and propagates it through the system using the performance-based probabilistic risk assessment (PRA) framework. Unlike traditional PRA approaches, it utilizes the power of Bayesian statistic to account for non-Boolean relationships and correlations among events at various levels of a network representation of the system. Bayesian updating facilitates evaluation of updated validation information as additional data from experimental observations or improved simulations is incorporated. PRA based framework assists in identifying risk-consistent events and critical path for appropriate allocation of resources to improve the validation.

Bayesian network application for the risk assessment of existing energy production units

The assessment of existing infrastructures in the energy sector is of great economic significance worldwide. Fossil power stations are reaching their design service life and rational decisions concerning extensions of service life, maintenance and replacements of devices should be based on updated information of the actual conditions of the energy devices and their components, and on cost-benefit analysis using risk analysis and probabilistic optimisation procedures.The contribution provides an integrated framework for probabilistic reliability and risk assessment of existing energy production units considering availability and human safety criteria. An extensive case study focused on risks of an energy production unit in a fossil power station is provided to support practical applications. A Bayesian network is thereby implemented to assess the risks of the selected production unit. Special emphasis is given to the input data consisting of failure rates obtained from recorded data and expert judgements. The influence of uncertainties in the considered performance indicators on the availability of the unit is analysed. It is shown that a reasonably simplified framework can provide a valuable assessment of the influence of individual devices and their components on availability and societal risk, identifying thus the major risk contributors.

Probabilistic risk assessment framework for structural systems under multiple hazards using Bayesian statistics

Conventional probabilistic risk assessment (PRA) methodologies (USNRC, 1983; IAEA, 1992; EPRI, 1994; Ellingwood, 2001) conduct risk assessment for different external hazards by considering each hazard separately and independent of each other. The risk metric for a specific hazard is evaluated by a convolution of the fragility and the hazard curves. The fragility curve for basic event is obtained by using empirical, experimental, and/or numerical simulation data for a particular hazard. Treating each hazard as an independently can be inappropriate in some cases as certain hazards are statistically correlated or dependent. Examples of such correlated events include but are not limited to flooding induced fire, seismically induced internal or external flooding, or even seismically induced fire. In the current practice, system level risk and consequence sequences are typically calculated using logic trees to express the causative relationship between events. In this paper, we present the results from a study on multi-hazard risk assessment that is conducted using a Bayesian network (BN) with Bayesian inference. The framework can consider statistical dependencies among risks from multiple hazards, allows updating by considering the newly available data/information at any level, and provide a novel way to explore alternative failure scenarios that may exist due to vulnerabilities.