首页期刊导航|Reliability engineering & system safety
期刊信息/Journal information
Reliability engineering & system safety
Elsevier Applied Science Publishers
Reliability engineering & system safety

Elsevier Applied Science Publishers

月刊

0951-8320

Reliability engineering & system safety/Journal Reliability engineering & system safetySCIISTP
正式出版
收录年代

    Letting losses be lessons: Human-machine cooperation in maritime transport

    Fan S.Shi K.Yang Z.Weng J....
    1.1-1.13页
    查看更多>>摘要：© 2024 The AuthorsNavigation safety has been a critical guarantee of global shipping, and it becomes more challenging given the increasing employment of advanced technologies and novel ship design in the era of Maritime Autonomous Surface Ships (MASS). The human-centred risk analysis of human-machine cooperation is scarce in general and emerging in maritime transport in specific. This paper aims to develop a new approach enabling the analysis of significant risk influencing factors (RIFs) in human-machine cooperation through an in-depth investigation of the occurred mistakes and violations in the cooperative operations of seafarers and machines in maritime transport. Its novelties consist of (1) a novel approach to analysing and quantifying the connectivity between humans and machines in safety-critical operations, (2) new integration of the frequency and impact of RIFs in the human-machine cooperation model, and (3) ultilisation of graph theory to generate a network to analyse critical human-machine RIFs and their interactions with the system. The connectivity analysis of RIFs is conducted through a weighted undirected network, showing the features of RIF connectivity accommodating the closed-loop system. The proposed novel approach, which combines the frequency and impact features to identify critical RIFs and analyses graphical features, will aid to realise the human-centred risk analysis for MASS. The findings make contributions for ship designers to rationalise the clustering design of function-based automation and training organisations to improve seafarer skills by rationally considering the identified risk-based human-machine cooperation features, and providing new competence schemes that can fit the demands of MASS in future.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Resilience evaluation of train control on-board system considering component failure correlations: Based on Apriori-Multi Layer-Copula Bayesian Network model

    Yu Y.Shuai B.Huang W.
    1.1-1.14页
    查看更多>>摘要：© 2024 Elsevier LtdThe failure of complex system components is an important factor affecting system resilience, and it is not only affected by their own basic life parameters, but may also be affected by the failure of other components. In order to investigate the impact of component failure correlations on the resilience evaluation of Train Control on Board System (TCOBS), we propose the Apriori-Multi Layer-Copula Bayesian Network (AMLCBN) model. Firstly, the definition and evaluation function of TCOBS component resilience are provided. Then, build a TCOBS Bayesian Network and perform hierarchical processing on the network to clarify the position of Copula functions in the Bayesian Network. The Copula function is used to evaluate the correlations among component failures, and the Copula Bayesian Network is used to infer TCOBS resilience. We use Apriori to calculate the correlation coefficient matrix in the Copula function. Finally, a case study is conducted by taking CTCS-3OBS as an example, the results show that among the components of TCOBS, BTM Ant has low resilience and high importance. Considering the correlation among component failures, the TCOBS resilience evaluation results will increase, and those components with higher importance will become more important, while those with lower importance will become less important.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    A closed-form continuous-depth neural-based hybrid difference features re-representation network for RUL prediction

    Li X.Hu Y.Wang H.Liu Y....
    1.1-1.11页
    查看更多>>摘要：© 2024Remaining Useful Life (RUL) prediction contributes to ensuring the reliability of mechanical systems and improving their maintenance plans. Recently, prediction methods based on deep learning have undergone rapid development. However, there are significant differences between multivariate monitoring sequences and utilizing a single model for feature extraction, which leads to reaching suboptimization. Additionally, focusing solely on a specific feature dimension usually imposes notable limitations on the model. This paper proposes a Closed-form Continuous-depth neural (CfC)-based hybrid difference Features Re-representation Network (CfC-F2RN) for RUL prediction. This method comprehensively utilizes hybrid difference features through two stages: initial feature representation and feature re-representation. In the initial feature representation stage, a Long Short-Term Memory (LSTM) is employed to capture the hidden state information of each time step in sequence data. Next, a novel attention mechanism is utilized to extract the hidden states and obtain a deep feature map. In the feature re-representation stage, the encoded features are re-represented using a decoder composed of the CfC model. Finally, the hidden state of the last LSTM unit in the encoder, serving as supplementary information, is combined with the decoded features to form a dual-latent feature representation of the mixed differential features. The prediction subnetwork is then adopted to accomplish RUL prediction. The superiority of CfC-F2RN is substantiated through benchmarking against established methods using the C-MAPSS dataset.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    A benchmark on uncertainty quantification for deep learning prognostics

    Basora L.Viens A.Olive X.Chao M.A....
    1.1-1.21页
    查看更多>>摘要：© 2024Reliable uncertainty quantification on RUL prediction is crucial for informative decision-making in predictive maintenance. In this context, we assess some of the latest developments in the field of uncertainty quantification for deep learning prognostics. This includes the state-of-the-art variational inference algorithms for Bayesian neural networks (BNN) as well as popular alternatives such as Monte Carlo Dropout (MCD), deep ensembles (DE), and heteroscedastic neural networks (HNN). All the inference techniques share the same inception architecture as functional model. The performance of the methods is evaluated on a subset of the large NASA N-CMAPSS dataset for aircraft engines. The assessment includes RUL prediction accuracy, the quality of predictive uncertainty, and the possibility of breaking down the total predictive uncertainty into its aleatoric and epistemic parts. Although all methods are close in terms of accuracy, we find differences in the way they estimate uncertainty. Thus, DE and MCD generally provide more conservative predictive uncertainty than BNN. Surprisingly, HNN achieve strong results without the added complexity of BNN. None of these methods exhibited strong robustness to out-of-distribution cases, with BNN and HNN methods particularly susceptible to low accuracy and overconfidence. BNN techniques presented anomalous miscalibration issues at the later stages of the system lifetime.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Fault Impulse Inference and Cyclostationary Approximation: A feature-interpretable intelligent fault detection method for few-shot unsupervised domain adaptation

    Zhang Q.Li S.Shi T.Xuan J....
    1.1-1.17页
    查看更多>>摘要：© 2024 Elsevier LtdWith the advancement of intelligent detection for rotating machinery, numerous domain adaptation methods have been devised to transfer detection knowledge from one source domain working condition to another target domain working condition, involving extensive transfer scenarios including labeled, few-shot labeled, and unlabeled target conditions. Yet, learning from sparsely labeled signals in the source domain working condition and transferring to unlabeled target conditions, termed few-shot unsupervised domain adaptation (FUDA), is closer to reality but almost unexplored. Diverging from the intuition of combining existing transfer and few-shot learning technologies, this paper pioneers a novel single learning principle focusing on the cyclostationary mechanism (CT) of fault signals. In its implementation, named cyclically enhanced cyclostationary variational autoencoder (CCTVAE), the CT principle motivates the encoder to infer domain-shared representations with fault impulses, and the decoder approximates the cyclostationary structure containing the clear fault and working condition information. Then, auxiliary samples for few-shot expansion are generated by adjusting cyclic parameters of the posterior distribution of representations. Experimentally, CCTVAE achieves commendable results on simulated and real fault datasets. Even for compound faults, domain-shared representations and generated auxiliary signals manifest interpretable fault-indicating spectral lines in the frequency domain, underscoring method reliability.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Probabilistic seismic risk analysis of electrical substations considering equipment-to-equipment seismic failure correlations

    Liang H.Xie Q.
    1.1-1.16页
    查看更多>>摘要：© 2024 Elsevier LtdWhen an earthquake occurs, electrical equipment in a substation exhibits a certain level of seismic failure correlation since they suffer similar ground motions and share similar structural characteristics. However, this equipment-to-equipment seismic failure correlation (E2ESFC) was neglected in previous substation-level probabilistic seismic risk analyses due to the lack of awareness and practical approach. To investigate the effect of different degrees of the E2ESFC on the substation seismic risk, an efficient method for considering partially correlated seismic failure was proposed. The concepts of “damage demand probability” and “damage capacity probability” were derived from the equipment's fragility curve. Then the partial correlation of equipment's capacity probabilities can be easily introduced and incorporated into the substation-level risk analysis through the combination of Copula functions and the Monte Carlo simulation. A case study on a real-world 220/110 kV substation using an equi-correlation model demonstrated that ignoring the E2ESFC among the same type of equipment will lead to an underestimate of the probability of seeing high seismic loss. Furthermore, a general method to assess the E2ESFC coefficients between equipment was also proposed, laying the foundation to facilitate applications of the introduced E2ESFC simulation method and to generate a more reliable system risk assessment result.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    A study of neural network-based evaluation methods for pipelines with multiple corrosive regions

    Zhang Z.Li S.Wang H.Qian H....
    1.1-1.17页
    查看更多>>摘要：© 2024 Elsevier LtdIn recent years, significant developments have been made in methods for assessing the remaining strength of corroded pipelines. However, existing methods have limitations as they mainly focus on the local impact of corrosion defects. This study explores evaluation methods using neural networks to predict the ultimate resistance of pipelines containing multiple corrosive regions. Firstly, based on the validated method, the study generates a dataset comprising 3,000 corroded pipeline models and pixelates the corrosion information of these models via digital images. Then, three neural network evaluation frameworks are constructed: a Multilayer Perceptron (MLP) using the overall corrosion matrix, an MLP based on corrosion feature parameters, and Convolutional Neural Networks (CNN) based on corrosion images. Following this, the study analyzes the relationship between various corrosion parameters and failure pressure, compares the training effectiveness of the three neural network methods, and validates the accuracy and applicability of the proposed approach. The results indicated that various corrosion features should be considered when evaluating corroded pipelines, particularly depth. In addition, all three neural network-based methods show improved applicability and reliability compared to traditional evaluation methods, with CNN-image having the highest evaluation accuracy (correlation coefficient = 0.9564, average error = 3.46%).
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Spatial network disintegration based on spatial coverage

    Deng Y.Wang Z.Xiao Y.Shen X....
    1.1-1.15页
    查看更多>>摘要：© 2024 Elsevier LtdThe problem of network disintegration, such as interrupting rumor spreading networks and dismantling terrorist networks, involves evaluating changes in network performance. However, traditional metrics primarily focus on the topological structure and often neglect the crucial spatial attributes of nodes and edges, thereby failing to capture the spatial functional losses. Here we first introduce the concept of spatial coverage to evaluate the spatial network performance, which is defined as the convex hull area of the largest connected component. Then a greedy algorithm is proposed to maximize the reduction of the convex hull area through strategic node removals. Extensive experiments verified that the spatial coverage metric can effectively quantify changes in the performance of spatial networks, and the proposed algorithm can maximize the reduction of the convex hull area of the largest connected component compared to genetic algorithm and centrality strategies. Specifically, our algorithm reduces the convex hull area by up to 30% compared to the best-performing strategy. These results indicate that the critical nodes influencing network performance are a combination of numerous peripheral spatial leaf nodes and a few central spatial core nodes. This study substantially enhances our understanding of spatial network robustness and provides a novel perspective for network optimization.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    DNN-metamodeling and fragility estimate of high-rise buildings with outrigger systems subject to seismic loads

    Xing L.Zhang P.Zhou Y.Gardoni P....
    1.1-1.16页
    查看更多>>摘要：© 2024This paper proposed deep neural networks (DNNs) for the dynamic response of high-rise buildings with one-outrigger systems under two types of seismic hazards. Using an existing database, the hyperparameters for the architecture of the DNNs are determined finding a trade-off between accuracy and complexity. The performance of the proposed DNNs is compared with two metamodels in the literature, a probabilistic demand model and a kriging metamodel. Partial dependence plot and SHapley Additive exPlanations are used for the explanation of the marginal effect of each feature on the predicted outcome of a neural network from the perspective of global and local agnostic. Considering the uncertainty in the input features, the DNNs are then used to formulate fragility estimates for example high-rise buildings with three types of outrigger systems. The model-agnostic analysis suggests that the DNN targeting the inter-story drift and top acceleration shows extremely high sensitivity to variation in the seismic hazard features, earthquake magnitude, and rupture distance. The fragility curves objectively quantify the reliability of buildings with each of the three outrigger systems and show the effectiveness of damped outrigger systems in reducing fragilities.
      原文链接:
    • NETL
    • NSTL
    • Elsevier

    Improving port state control through a transfer learning-enhanced XGBoost model

    Wang R.Yan R.Zhang M.Gong F....
    1.1-1.15页
    查看更多>>摘要：© 2024 Elsevier LtdWith the advancements in modern information technology, Port State Control (PSC) inspections, as a crucial measure to protect ship safety and the marine environment, are undergoing an intelligent transformation. This paper aims to streamline the selection process for inspecting high-risk ships by employing a data-driven model to predict the number of deficiencies in ships arriving at ports. A transfer learning-enhanced eXtreme Gradient Boosting (XGBoost) model is proposed by innovatively incorporating sample similarity calculations to adapt the model to the unique characteristics of the target port. This novel model enables the integration of relevant data from other ports, enhancing the predictive performance of the model to specific port conditions. Utilizing PSC inspection records from ports within the Tokyo Memorandum of Understanding (MoU) and choosing the port of Singapore as the target, numerical experiments demonstrate that the proposed model achieves improvements of 1.81 %, 6.08 %, and 3.60 % in the mean absolute error, mean squared error and root mean squared error, respectively, compared to the standard XGBoost model. Furthermore, across various sizes of training samples, the proposed model outperforms other machine learning models. This work may service as a significant step towards exploring the potential of developing data-driven models based on comprehensive data to assess the risk level of foreign ships arriving at ports, ameliorating the PSC inspection process by aiding PSC officers in identifying substandard ships more effectively.
      原文链接:
    • NETL
    • NSTL
    • Elsevier