Behavior Of Failures Research Articles

Influence of Foundation Stiffness on the Fragility Curves of a Concrete Gravity Dam Subjected to Near-Field Ground Motions Using Finite Element Method

A concrete gravity dam is being investigated to determine the fragility curves during near-field ground motions under various foundation stiffness conditions. This is accomplished using the finite element method for modeling the Latyan dam and its dam-reservoir-bed complex interaction system. The nonlinear constitutive model of concrete, plastic damage, is used to simulate the cyclic behavior of tensile and compressive failures in concrete. An acoustic element is used to model the propagation of compressive waves in the reservoir to account for the hydrodynamic effects of the lake reservoir. A total of 100 real and scaled accelerograms are used to simulate near-filed seismic loading in eight different concrete-to-foundation stiffness ratios. Based on the peak ground acceleration (PGA) of the earthquake, numerical models are applied to obtain the probability density function (PDF) and cumulative distribution function (CDF) for lognormal distributions and to determine and analyze the fragility curves of the dam. According to the findings, the seismic performance of the dam is improved for stiffness ratios of 0.75, 1, and 1.5, and the best seismic performance is determined for stiffness ratios of 1. Concrete dams with stiffness ratios of 0.75, however, perform better when the peak ground acceleration is greater than 0.74g compared to those with stiffness ratios of 1. DOI: https://doi.org/10.52783/pst.502

Critical behavior of cascading failures in overloaded networks.

While network abrupt breakdowns due to overloads and cascading failures have been studied extensively, the critical exponents and the universality class of such phase transitions have not been discussed. Here, we study breakdowns triggered by failures of links and overloads in networks with a spatial characteristic link length ζ. Our results indicate that this abrupt transition has features and critical exponents similar to those of interdependent networks, suggesting that both systems are in the same universality class. For weakly embedded systems (i.e., ζ of the order of the system size L) we observe a mixed-order transition, where the order parameter collapses following a long critical plateau. On the other hand, strongly embedded systems (i.e., ζ≪L) exhibit a pure first-order transition, involving nucleation and the growth of damage. The system's critical behavior in both limits is similar to that observed in interdependent networks.

Assessment of QOS indicators of a network with UDP and TCP traffic under a node peak load mode

The object of research is Markov models of network nodes with UDP (User Datagram Protocol) and TCP (Transmission Control Protocol) traffic and their differences. The task solved is the lack of Markov models of network nodes describing the behavior of TCP traffic from the point of view of packet retransmissions and packet delivery guarantees. Markov models of network nodes describing traffic behavior with guaranteed packet delivery have been further advanced. Given the comparison of the models, the differences from the classic models serving TCP traffic were shown, for each packet flow, an additional dimensionally was added to the graph of states and transitions, which takes into account the retransmission of a lost packet. The comparison graph shows similar behavior of queue length and packet loss for both types of traffic. But the nature of the curves is different. With TCP traffic, packet loss can exceed 5 percent. In addition, lost packets must be retransmitted, which increases the load on the network node. More failures and packet queue lengths at a network node during peak load typically occur with TCP traffic compared to UDP traffic. At peak load, the difference in service failures can reach 20–30 percent. The main reason is that TCP uses flow control and rate-limiting mechanisms to avoid network congestion and ensure efficient data transfer between nodes. The Markov model of TCP traffic requires an additional dimensionally on the graph of states and transitions, which affects the behavior of queues and packet failures. The investigated problem was solved due to the universality and diversity of the mathematical apparatus of Markov mass service systems. The results could be used in network modeling software products for building and reengineering the topology of electronic communications networks at enterprises and organizations

Fault Diagnosis using eXplainable AI: A transfer learning-based approach for rotating machinery exploiting augmented synthetic data

Due to the growing interest for increasing productivity and cost reduction in industrial environment, new techniques for monitoring rotating machinery are emerging. Artificial Intelligence (AI) is one of the approaches that has been proposed to analyze the collected data (e.g., vibration signals) providing a diagnosis of the asset’s operating condition. Three major problems hinder the application of AI models in industrial environments, being the motivation for the research: (i) impossibility or long time to obtain a sample of all operational conditions (since faults rarely happen), (ii) high cost of experts to label all acquired data (in a supervised scenario) and (iii) lack of interpretability of the models (black-boxes). To overcome these problems, a new generic and interpretable approach for classifying faults in rotating machinery based on transfer learning from augmented synthetic data to real rotating machinery is here proposed, namely FaultD-XAI (Fault Diagnosis using eXplainable AI). To provide scalability using transfer learning, synthetic vibration signals are created mimicking the characteristic behavior of failures in operation. The application of Gradient-weighted Class Activation Mapping (Grad-CAM) with 1D Convolutional Neural Network (1D CNN) allows the post hoc interpretation of results, supporting the user in decision making and increasing diagnostic confidence. The proposed approach not only obtained promising diagnostic performance but was also able to learn characteristics used by experts to identify conditions in a source domain and apply them in another target domain. The experimental results obtained on three datasets containing different mechanical faults suggest the method offers a promising approach on exploiting transfer learning, synthetic data and explainable artificial intelligence for fault diagnosis. Lastly, to guarantee reproducibility and foster research in the field, the developed dataset is made publicly available.

Alatent‐factor self‐excitingpoint process for software failures

AbstractSoftware debugging is the process of detecting and removing bugs during software development. Although the intent of modifications to the software is to remove bugs, one cannot rule out the possibility of introducing new bugs as a result of these modifications. We consider a self‐exciting point process, which can incorporate the case of reliability deterioration due to the potential introduction of new bugs to the software during the development phase. In order to account for the unobservable process of introducing bugs, latent variables are incorporated into the self‐exciting point process models. The models are then applied to two data sets in software reliability and additional insights that can be obtained from these models are discussed. Our results suggest that the self‐exciting processes with latent factors perform better than the standard point process models in describing the behavior of software failures during the debugging process.

A Pareto–Poisson process model for intermittent debris flow surges developing from slope failures

Many debris flows originate from shallow slope failures over the source area of valley and move in form of successive surges. In order to understand the behavior of such failures, we conducted seven sets of experiments of rainfall-induced soil failures in a 4 × 8 m spot on a slope under different rainfall intensities ranging from 12 to 60 mm/h, according to the local rainfall conditions in the valley of Jiangjia Gully, and this work is based on the experimental data. It is found that the failures are realization of a Poisson process with failure volume satisfying the Pareto distribution, thus the failure process can be simulated using a combinative Pareto-Poisson process (PPP) model. The PPP model incorporates a Poisson process determining the failure time, and a Pareto distribution characterizing the fluctuation of failure volume. The developing course of a debris flow surge is reduced to the hierarchic evolution of the failures from their source to the mainstream, which amounts to a cascade thinning process of the Poisson sequence in the PPP model framework. Given the controlling parameters (i.e., the Poisson intensity, Pareto distribution parameters, and thinning ratio), the model is capable of predicting surge sequences that agree well with the monitored surges in terms of time interval, magnitude fluctuation, and probability distribution. This study represents the first attempt to formulate a stochastic framework for debris flows developing from source to mainstream, which applies to debris flows fed by discontinuous material supplies from shallow slope failures.

A systematic method for the optimization of gas supply reliability in natural gas pipeline network based on Bayesian networks and deep reinforcement learning

This study proposes a method based on Bayesian networks (BNs) to optimize the reliability of gas supply in natural gas pipeline networks. The method integrates probabilistic safety analysis with preventive maintenance to achieve the targets of minimizing gas shortage risk and reducing maintenance costs. For this, the tasks of unit failure probability calculation, system maximum supply capacity analysis, gas supply reliability assessment and system maintenance planning are performed. A stochastic capacity network model is coupled with a Markov model and graph theory to generate the state space of the pipeline network system. BN, is then, proposed as the modeling framework to describe the stochastic behavior of unit failures and customer gas shortage. The system maintenance problem is converted into a Markov decision process (MDP), and solved by using deep reinforcement learning (DRL). The effectiveness of the proposed method is validated on a case study of a European gas pipeline network. The results show that the proposed method outperforms others in identifying optimal maintenance strategies. The DRL-optimized maintenance strategy is capable of responding to a dynamic environment through continuous online learning, considering the randomness of the unit failures and the uncertainty in gas demand profiles.

Electrical Submersible Pump Design Enhancements for Hydrogen Sulfide Harsh Environments

Summary Electrical submersible pumping (ESP) remains the preferred artificial lift method for high rate production when technically viable. ESP, on the other hand, is sensitive to downhole conditions and pumped fluid. Sour fields, in particular, are considered as a major challenge for producing facilities and well completion elements. Reservoirs producing fluids with hydrogen sulfide (H2S) present a special challenge to ESP systems. This paper uses ESP field observations and pulled equipment findings from many dismantle inspection and failure analyses (DIFAs). The findings confirmed H2S behavior and root causes of electrical and mechanical failures within multiple ESP components. The outcome of these investigations and the recommended system upgrades to enhance ESP reliability in corrosive environments will be illustrated. Critical ESP system materials will deteriorate and fail when subjected to sour environments. H2S can penetrate the pump’s cable insulation, attack the copper, and react to form copper sulfide, resulting in electrical failure. It can also permeate the seal bags and o-rings, diffuse in the seal dielectric oil, and attack the bronze and copper components in the seal and the motor. To improve reliability, a new version of motor lead extension (MLE) using three individually armored connectors and a seal with H2S sacrificial anode scavenger inside each chamber were introduced. The improved design encapsulated the insulated conductors individually within metal tubes made of high nickel alloy. The tubes can be terminated individually at the motor and above the production packer with proven swage type connectors. By utilizing high nickel alloy tubes as barriers against H2S and removing all connections below the packer, the H2S effect has been eliminated. On the other hand, the seal with H2S passive scavenger will retain most of the H2S in the dielectric oil before it reaches the motor. These novelty technologies resulted in a threefold improvement, leading to longer up time, less workover jobs, and more sustainable production.

Man-Machine Modeling and Safety Analysis Method for Landing Process Based on SysML and Simulink

The large-scale system’s mission process and the interaction within the system are becoming more complex with the improvement of the integration and intelligence, and the complex interactions among multiple factors, such as the unsafe behavior of personnel, equipment failures, and environmental interference, make safety analysis a greater challenge. Aimed at the safety of carrier aircraft, an integrated system modeling and safety analysis method for aircraft landing process were proposed based on System Modeling Language (SysML) and Simulink. First, the SysML models were built according to the analysis of the mission process, including system structure and behavior process, using multiple diagrams. Second, the SysML models were transformed into and integrated with the Simulink platform to build entity models with continuous dynamic characteristics and to perform safety analysis through simulation. Finally, an example of aircraft attitude control during landing was given to demonstrate the proposed method, and the safety states were analyzed and assessed under different disturbance conditions.

Contact-Fatigue Resistance of PM Steel with Hard Coatings

Hard coatings applied by the Physical Vapor Deposition method often provide very good mechanical properties, especially when applied to metal parts that are mentioned to withstand certain level of wear. In this study, TiN coating combined with deep rolling were applied to the investigated steel samples, prepared by powder metallurgy and subjected to contact-fatigue stress. Computational analysis of stress states in samples by the finite element method helped to reveal the behavior and formation of fatigue failures when loading samples. The results, processed in the form of fatigue curves, probabilistic Weibull curves using metallography and electron microscopy, showed a positive effect of the used coatings and their combination with other surface treatments on the contact-fatigue strength of the examined samples. Microscopic study also showed the different mechanisms of crack formation and of crack propagation rate due to contact loading of material with a laser hardened surface, which has an obvious impact on material lifetime.

Robustness of spontaneous cascading dynamics driven by reachable area

Different from the previous research framework of cascading failure driven by intentional attack or random failure, this paper considers the occurrence of cascading failure from a new perspective, i.e., the spontaneous phenomenon of cascading failure. When the network changes from the low peak period to the high peak period, the global distribution of load changes, which leads to the cascading failure of the network. This phenomenon is called the spontaneous phenomenon of cascading failure. We introduce the concept of reachable region and construct a new cascading failure model, which mainly focus on how the dynamic changes of reachable area drive the dynamic behavior of cascade failures. Through simulation on two infrastructure networks, we found that: (1) The size of reachable area is positively correlated with network robustness. The larger the reachable area is, the smaller the load fluctuation of the whole network is, and the cascading failures are more easily alleviated and the network robustness is higher. But when the edge load capacity increases, the robustness of the network is not necessarily improved, which indicates the existence of capacity paradox. (2) The size of reachable area is negatively correlated with network capacity redundancy. The larger the reachable area is, the lower the capacity redundancy of the whole network will be and the less network resources will be wasted. (3) Combining with the network topology and the phenomenon of “key edge” we find that the edge with less betweenness centrality usually needs to invest higher resource protection.

AC Cascading Failure Model for Resilience Analysis in Power Networks

Cascading failures are one of the main mechanisms causing widespread blackouts of power networks. Models simulating the behavior of cascading failures are widely used in the literature to understand fault propagation and investigate effective mitigation strategies. However, there is a lack of validated models that address the specific requirements of resilience analysis in power networks and that are computationally fast and converge reliably for very large contingency sizes that may occur under extreme events. This article presents a novel comprehensive ac cascading failure model particularly designed for resilience analysis in power networks. The model is capable to deal with large contingency sizes, it is computationally efficient in large networks and integrates seamlessly with established resilience metrics. It incorporates dynamic phenomena and protection mechanisms using static representations. The model is verified following the recommendations by the IEEE PES working group on cascading failures using internal validation, sensitivity analysis, and comparison to historical outage data. Furthermore, an analysis of the impact of different contingency sizes and the dependency of cascades on network loading level, are given to illustrate some applications of the model and to highlight its capabilities.

Reliability-constrained dynamic transmission expansion planning considering wind power generation

This paper proposes a two-stage framework to solve the long-term transmission network expansion planning (TNEP) problem ensuring user-defined reliability levels and wind curtailment over the planning horizon. In the first stage, the static TNEP (S-TNEP) problem is solved to define where a set of new lines must be installed at the end of planning horizon. In the second stage, a multistage procedure is performed to solve the dynamic TNEP (D-TNEP) in order to determine the moment that each transmission line should be built. Both S-TNEP and D-TNEP are decomposed into investment decision and performance assessment through a bi-level scheme based on Benders decomposition; thus, the reliability is considered and evaluated over the planning process. The reliability and performance indexes are obtained through the non-chronological Monte Carlo simulation considering the random behavior of transmission lines failures, load fluctuations and uncertainties over wind availability. The loss of wind probability performance index is introduced to measure the probability of wind curtailment in each connection point of the system, and it is used to prevent wind farms from being underused. The proposed methodology is tested in modified versions of the 24-bus IEEE reliability test system and the IEEE 118-bus test system.

Universal behavior of cascading failures in interdependent networks.

Catastrophic and major disasters in real-world systems, such as blackouts in power grids or global failures in critical infrastructures, are often triggered by minor events which originate a cascading failure in interdependent graphs. We present here a self-consistent theory enabling the systematic analysis of cascading failures in such networks and encompassing a broad range of dynamical systems, from epidemic spreading, to birth-death processes, to biochemical and regulatory dynamics. We offer testable predictions on breakdown scenarios, and, in particular, we unveil the conditions under which the percolation transition is of the first-order or the second-order type, as well as prove that accounting for dynamics in the nodes always accelerates the cascading process. Besides applying directly to relevant real-world situations, our results give practical hints on how to engineer more robust networked systems.

Comparative study on the transversal/lengthwise thermal failure propagation and heating position effect of lithium-ion batteries

Because of the multi-layer structure and the diversified connection modes of most battery modules, systematically investigating the thermal failure propagation principles and mechanisms of lithium-ion batteries is critical to provide early warnings and protection for thermal runaway. In this work, a series of experimental studies and mathematical deductions were conducted to investigate the propagation behavior of thermal failures within two types of cells under various heating modes, and their heat transfer mechanisms were analyzed. In general, the thermal runaway phenomenon in Li (Ni1/3Co1/3Mn1/3) O2 cells is more severe than that in LiCoO2 cells. For different heating modes, lengthwise thermal failure propagation is more unlikely to occur in comparison with transversal thermal failure propagation; however, the former involves a more violent combustion. For different heating positions, heating near the positive pole results in the most violent phenomena. Additionally, a higher Tmax of 185.6 °C was obtained via middle heating in comparison with that obtained via heating on the poles. The temperature rising rate also varied, taking 1619, 1578, and 1699 s for the temperature to rise from 20 °C to 168.9 °C, 185.6 °C, and 173.4 °C through bottom, middle, and top heating, respectively. These phenomena were consequently ascribed to the different heat transfer rates along different directions inside the cells, including transversal/lengthwise propagation, and positive-pole-directional/negative-pole-directional propagation. These encouraging results may raise concerns about developing more precise and suitable surveillance and control measures to further enhance the thermal safety performance of cells/modules from both external and internal perspectives.

Metallography of an Underground Oil Pipeline

The purpose of this article is to show a metallographic analysis of an underground pipeline taken out of operation upon failure. The pipeline had an 8.89 cm (3.5”) diameter and a 7 mm thickness. The study was based on a 45 cm long pipe sample, visibly and entirely corroded, with a fish-mouth crack along its length. The work contributes to finding new ways to prevent structural failure, which has high-impact consequences from the point of view of production, damage to property, pollution, and risks to human live. Through this analysis the knowledge on behavior of failures in terrestrial ducts has been extended. Development of the research included metallographic, chemical, and mechanical tests on the sample in order to know the composition of the material, its strength and its physical conditions upon taking it out of operation. After the analysis of the laboratory tests, the physical and chemical features were compared to existing national and international regulations, which allowed a specific characterization of the conditions of the sample. In accordance with the regulations, the grade of the pipe was between ×65 and ×70. Tensile testing was carried out to obtain mechanical properties in order to corroborate the grade of pipeline steel and complement the metallographic analysis.

A reliability-based approach to maximize availability in a water supply system

In some regions where the water supply system is deficient and unable to supply all consumers simultaneously, the appropriate maintenance management allows to improve the system's availability. As the maintenance activities, either preventive or corrective, represents system's service disruption, non-accurate maintenance schedules may reduce system's availability. In such context, availability has to be optimized even if it results in higher costs due to the water supply system deficiencies to increase the hours of water supply. This paper presents a real situation that is found in many undeveloped countries and proposes a model to maximize the availability of a water supply system. From a reliability perspective, the behavior of system failures is studied and subsequently the availability is optimized considering preventive and corrective maintenance actions. To illustrate the approach proposed in this paper, a case study is presented with an application of the model in a system supplied by a water well, allowing quantify the system's reliability and reduce its downtime.

Belief Universal Generating Function Analysis of Multi-State Systems Under Epistemic Uncertainty and Common Cause Failures

Because of the complexity of engineering systems, and the fact that insufficient data are only available to obtain the precise state probability of components, an extended universal generating function (UGF) based on belief function theory is introduced in this paper to conduct the reliability analysis of multi-state systems (MSSs) with epistemic uncertainty. The behavior of common cause failures (CCFs) is further incorporated, and the occurrence probability of CCFs is evaluated using a weighted impact vector method. A numerical example is used to illustrate how the proposed method works. In addition, a global optimization method is used to obtain the truth interval of the system reliability, and the results are compared with those obtained by using some existing methods. The case study shows that the belief UGF method can effectively avoid the interval expansion problem and the overestimation problem involved in the interval UGF method, and the proposed method can be used to provide a reliable way to evaluate the reliability of MSSs with interval data and CCFs.

Spatial correlation analysis of cascading failures: congestions and blackouts.

Cascading failures have become major threats to network robustness due to their potential catastrophic consequences, where local perturbations can induce global propagation of failures. Unlike failures spreading via direct contacts due to structural interdependencies, overload failures usually propagate through collective interactions among system components. Despite the critical need in developing protection or mitigation strategies in networks such as power grids and transportation, the propagation behavior of cascading failures is essentially unknown. Here we find by analyzing our collected data that jams in city traffic and faults in power grid are spatially long-range correlated with correlations decaying slowly with distance. Moreover, we find in the daily traffic, that the correlation length increases dramatically and reaches maximum, when morning or evening rush hour is approaching. Our study can impact all efforts towards improving actively system resilience ranging from evaluation of design schemes, development of protection strategies to implementation of mitigation programs.

Network Formation in the Presence of Contagious Risk

There are a number of domains where agents must collectively form a network in the face of the following trade-off: each agent receives benefits from the direct links it forms to others, but these links expose it to the risk of being hit by a cascading failure that might spread over multistep paths. Financial contagion, epidemic disease, the exposure of covert organizations to discovery, and electrical power networks are all settings in which such issues have been articulated. Here we formulate the problem in terms of strategic network formation, and provide asymptotically tight bounds on the welfare of both optimal and stable networks. We find that socially optimal networks are, in a precise sense, situated just beyond a phase transition in the behavior of the cascading failures, and that stable graphs lie slightly further beyond this phase transition, at a point where most of the available welfare has been lost. Our analysis enables us to explore such issues as the trade-offs between clustered and anonymous market structures, and it exposes a fundamental sense in which very small amounts of “over-linking” in networks with contagious risk can have strong consequences for the welfare of the participants.