Dynamic Object-oriented Bayesian Network Research Articles

Safety analysis of blended hydrogen pipelines using dynamic object-oriented bayesian network

Energy demand is increasing rapidly while traditional fossil fuel resources are depleting. Hydrogen is considered a promising alternate fuel. The safety of hydrogen transportation is a key issue that warrants greater attention, primarily to determine how existing pipelines of natural gas transportation can be used for hydrogen transportation. This study has proposed a safety analysis model using the physics and mechanistic approach of hydrogen-associated degradation. A dynamic Object-Oriented Bayesian network (OOBN) model is proposed to study the mechanisms and physics of failure with hydrogen-associated degradation. The proposed model is explained using an industrial case study. Results show low-strength steel is less susceptible to hydrogen-associated degradation than high-strength steel pipes. Moisture, internal stress, and loss of metal ductility are key parameters that require detailed attention. The results of dynamic OOBN also indicate that hydrogen release probability follows a non-linear behavior dependent on the type of metal and operating conditions. The study identifies that controlling the corrosion rate can help ensure the safety of blended hydrogen pipelines.

Research on aging-related degradation of control rod drive system based on dynamic object-oriented Bayesian network and hidden Markov model

The control rod drive system is critical to the reactor's reliable operation. The performance of its control system and mechanical system will gradually deteriorate because of operational and environmental stresses, thus increasing the reactor's operational risk. Currently there are few researches on the aging-related degradation of the entire control rod drive system. Because it is difficult to quantify the effect of various environmental stresses and establish an accurate physical model when multiple mechanisms superimposed in the degradation process. Therefore, this paper investigates the aging-related degradation of a control rod drive system by integrating Dynamic Object-Oriented Bayesian Network and Hidden Markov Model. Uncertainties in the degradation of the control system and mechanical system are addressed by using fuzzy theory and the Hidden Markov Model respectively. A system which consists of eight control rod drive mechanisms divided into two groups is used to demonstrate the method. The aging-related degradation of the control rod drive system is analyzed by the Bayesian inference algorithm based on the accelerated life test data, and the impact of different operating schemes on the system performance is also investigated. Meanwhile, the components or units that have major impact on the system's performance are identified at different operational phases. Finally, several essential safety measures are suggested to mitigate the risk caused by the system degradation.

Early warning method for overseas natural gas pipeline accidents based on FDOOBN under severe environmental conditions

The completion and operation of transnational long-distance oil and gas pipelines will not only alleviate shortages of oil and gas resources in China, but also lead to new development opportunities in the countries along their routes. However, the frequent occurrence of severe environmental disasters has led to several uncertainties regarding the long-term safe and stable operation of overseas pipelines, including pipeline fracture, fire, explosions, etc. In this paper, the fuzzy dynamic object-oriented Bayesian network (FDOOBN) theory was introduced to establish an early warning method for overseas natural gas pipeline accidents under harsh environmental conditions. Firstly, for the harsh environmental conditions (lightning, rain, and wind) at pipeline laying stations, accident scenarios under a single harsh environmental condition and the combination of multiple harsh environmental conditions were constructed. The object-oriented concept was adopted to modularize the station system and equipment, and the dynamic Bayesian network (DBN) model of each subsystem of the station was established. Then, the conditional probability parameters in the model were determined by the fuzzy mathematical method. The DBN model of each subsystem and the dynamic object-oriented Bayesian network (DOOBN) model of the entire station system based on the process flow of the station and the object-oriented concept were introduced in turn to establish the fuzzy dynamic Bayesian network (FDBN) model and the FDOOBN model respectively. The dynamic early warning system for station risk under harsh environmental conditions was finally realized. Finally, the prediction errors of environmental parameters, such as meteorological conditions, were introduced to modify the reliability of the model. The results show that compared with the traditional model, the error-corrected FDOOBN model not only has a better performance in simplifying the modelling process and fully integrating expert experience, but also has an increase in dynamic warning range, further improving the reliability of accident warnings.

Dynamic ecological risk modelling of hydrocarbon release scenarios in Arctic waters

The Arctic is an ecologically diverse area that is increasingly vulnerable to damages from oil spills associated with commercial vessels traversing newly open shipping lanes. The significance of such accidents on Arctic marine habitats and the potential for recovery can be examined using ecological risk assessment (ERA) coupled with a dynamic object-oriented Bayesian network (DOOBN). A DOOBN approach is useful to represent the probabilistic relationships inherent in the interactions between key events associated with an oil spill, including oil dispersion from the source, ice-oil slick interactions, seawater-oil slick formation, sedimentation, and exposures to different aquatic life. From such analysis, a probabilistic cost analysis can be performed to examine the theoretical cost of habitat services lost and restored. The application of an ERA-DOOBN model to assess oil spills in the Arctic is demonstrated using a case study. The utility of the model output for determining habitat restoration costs and developing policy guidelines for ecological response measures in the Arctic is also discussed.

Assessment of flood risk in Mediterranean catchments: an approach based on Bayesian networks

National and international technical reports have demonstrated the increase of extreme event occurrences which becomes more dangerous in coastal areas due to their higher population density. In Spain, flood and storm events are the main reasons for compensation according to the National Insurance Consortium. The aim of this paper is to model the risk of flooding in a Mediterranean catchment in the South of Spain. A hybrid dynamic object-oriented Bayesian network (OOBN) was learnt based on mixture of truncated exponential models, a scenario of rainfall event was included, and the final model was validated. OOBN structure allows the catchment to be divided into five different units and models each of them independently. It transforms a complex problem into a simple and easily interpretable model. Results show that the model is able to accurately watch the evolution of river level, by predicting its increase and the time the river needs to recover normality, which can be defined as the river resilience.

Resilience modeling of engineering systems using dynamic object-oriented Bayesian network approach

Resilience is defined as the ability of a system to efficiently recover from a disruptive event. For an engineered system, it is dependent on the system performance and time. In this study, a new resilience assessment methodology is proposed. The methodology considers an engineered system to be characterized with three resilience capacities of absorption, adaptation, and restoration. These capacities are modeled as a function of time using dynamic object-oriented Bayesian network formalism. The application of the methodology is demonstrated using the Fukushima Daiichi nuclear power accident. The analysis showed that the adaptive capacity contributed most to the system resilience, followed by the absorptive capacity and the restorative capacity. The applicability of the proposed formalism has demonstrated that the concept of resilience is highly valuable for engineering design, operation, and emergency planning.

A comprehensive method for safety management of a complex pump injection system used for shale-gas well fracturing

Abstract A pump injection system used in the shale-gas well fracturing process is subjected to various adverse factors during its service, such as high pressures of up to 105 MPa and a large displacement, leading to a high failure rate and a rapid degradation in system performance. To ensure the safety and reliability of such a system, a comprehensive safety management method based on a dynamic object-oriented Bayesian network (DOOBN) is proposed in this article. The approach provides a framework that integrates a system function model, causal model, system behaviour model, and online fault diagnosis model with a remaining life prediction model, to characterise the behaviours in a complex system, such as fault propagation and system degradation. This method could achieve fault diagnosis and also predict the degradation trend of critical components and system performance in the long term, starting from the current system state. The application of the integrated safety management approach to the specific example of the pump injection system demonstrates how each phase of the presented method contributes to the achievement of fault diagnosis and residual life prediction in a systematic and holistic way. It is shown that the proposed model is a reasonable starting point for forecasting the remaining life of pump injection systems. This approach could be integrated into a real-time safety warning device for field application.

Reliability Evaluation Methodology of Complex Systems Based on Dynamic Object-Oriented Bayesian Networks

A novel reliability evaluation methodology of complex systems is proposed using dynamic object-oriented Bayesian networks (DOOBNs). This modeling methodology consists of two main phases, namely, construction phases for object-oriented Bayesian networks (OOBNs) and for DOOBNs. In the first phase, the network fragments with similar structures and parameters are divided into classes; these classes are then encapsulated. The construction of OOBNs is completed according to the relationship among the encapsulated classes. In the second phase, every fragment of the dynamic Bayesian networks that was constructed by the first phase is encapsulated as a class called DOOBN. The construction of DOOBNs is completed according to the relationship among the time fragments. The accuracy of this methodology is validated using the all-series, all-voting, voting-after-series, series-after-voting, parallel-after-series, and series-after-parallel systems. This methodology is further illustrated and verified by using deepwater blowout preventer system and can model the system from global to local levels, thereby effectively reducing modeling difficulty and adopting efficient arithmetic reasoning algorithms.

A road map for developing and applying object-oriented bayesian networks to “WICKED” problems

Wildlife managers, conservation-based authorities and NGOS are often tasked to provide management guidelines concerning “wicked” problems and Ecosystem Services (ESS) with scarce data and limited resources. More often than not, this dilemma is further intensified by competing objectives and the need for economic profitability. A range of quantitative and qualitative models are often applied to model ecosystem services. However, lack of quantifiable data remains an obstacle. We provide a brief overview of Bayesian Networks (BNs), Object Oriented Bayesian Networks (OOBNs), Dynamic Bayesian Networks (DBNs), and Dynamic Object Oriented Bayesian Networks (DOOBNs) construction and a flow chart highlights their utility. OOBNs are a semi-quantitative modelling approach that uses Bayes probability theorem to represent ecological complexities and conservation concerns. Its transparency allows it to act as a conceptual system and a decision support tool. And yet, Bayesian networks are not commonly applied to the ecological discipline because they are computationally expensive. We promote the notion that sustainable management is advanced by providing biology-based decision-making networks to address wicked problems. First, we provide a brief reasoning for the applicability of Bayes theorem to wicked challenges. Then, we demonstrate the efficiency of Bayesian modelling by applying it to the case study of the black wildebeest (Connochaetes gnou) and blue wildebeest (Connochaetes taurinus) hybridization concern in South Africa. Empirical data, in addition to expert explicit understanding of uncertainties was utilized to infer the probability that specific ecological, biological, and economic parameters, and their relationships, facilitate hybridization. We found the following management variables to be key to impact the probability of hybridization: (i) blue wildebeest male to black wildebeest male ratio, (ii) land cover (tree savanna vs. grassland), and (iii) spatial connectivity (i.e., fences). By quantifying ecological uncertainty, OOBN illuminates decision-making tradeoffs and serves to promote informed decision making either during the risk assessment stage or the management impact analysis stage. We offer a SWOT analysis (Strength, weaknesses, opportunities, and threats) and put forth a step-by-step user friendly roadmap by which OOBNs may be applied to solve similar “wicked” problems. We identify three primary phases: phases: (1) a priori data generation, (2) model development, and (3) model validation and verification. Our suggested step-by-step framework of OOBN construction and validation is of global relevance; it is designed to model similar “wicked” problems worldwide.

Constructing a clinical decision-making framework for image-guided radiotherapy using a Bayesian Network

A decision-making framework for image-guided radiotherapy (IGRT) is being developed using a Bayesian Network (BN) to graphically describe, and probabilistically quantify, the many interacting factors that are involved in this complex clinical process. Outputs of the BN will provide decision-support for radiation therapists to assist them to make correct inferences relating to the likelihood of treatment delivery accuracy for a given image-guided set-up correction. The framework is being developed as a dynamic object-oriented BN, allowing for complex modelling with specific subregions, as well as representation of the sequential decision-making and belief updating associated with IGRT. A prototype graphic structure for the BN was developed by analysing IGRT practices at a local radiotherapy department and incorporating results obtained from a literature review. Clinical stakeholders reviewed the BN to validate its structure. The BN consists of a sub-network for evaluating the accuracy of IGRT practices and technology. The directed acyclic graph (DAG) contains nodes and directional arcs representing the causal relationship between the many interacting factors such as tumour site and its associated critical organs, technology and technique, and inter-user variability. The BN was extended to support on-line and off-line decision-making with respect to treatment plan compliance. Following conceptualisation of the framework, the BN will be quantified. It is anticipated that the finalised decision-making framework will provide a foundation to develop better decision-support strategies and automated correction algorithms for IGRT.

Complex system reliability modelling with Dynamic Object Oriented Bayesian Networks (DOOBN)

Nowadays, the complex manufacturing processes have to be dynamically modelled and controlled to optimise the diagnosis and the maintenance policies. This article presents a methodology that will help developing Dynamic Object Oriented Bayesian Networks (DOOBNs) to formalise such complex dynamic models. The goal is to have a general reliability evaluation of a manufacturing process, from its implementation to its operating phase. The added value of this formalisation methodology consists in using the a priori knowledge of both the system's functioning and malfunctioning. Networks are built on principles of adaptability and integrate uncertainties on the relationships between causes and effects. Thus, the purpose is to evaluate, in terms of reliability, the impact of several decisions on the maintenance of the system. This methodology has been tested, in an industrial context, to model the reliability of a water (immersion) heater system.