Deep Auto-Encoder and Deep Forest-Assisted Failure Prognosis for Dynamic Predictive Maintenance Scheduling

Prognostics and health management (PHM) with failure prognosis and maintenance decision-making as the core is an advanced technology to improve the safety, reliability, and operational economy of engineering systems. However, studies of failure prognosis and maintenance decision-making have been conducted separately over the past years. Key challenges remain open when the joint problem is considered. The aim of this paper is to develop an integrated strategy for dynamic predictive maintenance scheduling (DPMS) based on a deep auto-encoder and deep forest-assisted failure prognosis method. The proposed DPMS method involves a complete process from performing failure prognosis to making maintenance decisions. The first step is to extract representative features reflecting system degradation from raw sensor data by using a deep auto-encoder. Then, the features are fed into the deep forest to compute the failure probabilities in moving time horizons. Finally, an optimal maintenance-related decision is made through quickly evaluating the costs of different decisions with the failure probabilities. Verification was accomplished using NASA’s open datasets of aircraft engines, and the experimental results show that the proposed DPMS method outperforms several state-of-the-art methods, which can benefit precise maintenance decisions and reduce maintenance costs.

Applications of FFTA–HFACS for Analyzing Human and Organization Factors in Electric Misoperation Accidents

Human and organizational factors (HOFs) play an important role in electric misoperation accidents (EMAs), but research into the reliability of human factors is still in its infancy in the field of EMAs, and further investment in research is urgently required. To analyze the HOFs in EMAs, a hybrid method including the Human Factors Analysis and Classification System (HFACS) and fuzzy fault tree analysis (FFTA) was applied to EMAs for the first time in the paper. HFACS is used to identify and classify the HOFs with 135 accidents, reorganized as basic events (BEs), intermediate events (IEs), and top event (TE), and develop the architecture of fault tree (FT). Fuzzy aggregation is employed to address experts’ expressions and obtain the failure probabilities of the BEs and the minimal cut sets (MCSs) of the FT. The approach generates BEs failure probabilities without reliance on quantitative historical failure statistics of EMAs via qualitative records processing. The FFTA–HFACS model is applied for quantitative analysis of the probability of failure of electrical mishaps and the interaction between accident risk factors. It can assist professionals in deciding whether and where to take preventive or corrective actions and assist in knowledgeable decision-making around the electric operation and maintenance process. Finally, applying this hybrid method to EMAs, the results show that the probability of an EMAs is 1.0410 × 10−2, which is a risk level that is likely to occur and must be controlled. Two of the most important risk factors are habitual violations and supervisory violation; a combination of risk factors of inadequate work preparation and paralysis, and irresponsibility on the part of employees are also frequent errors.

Reliability-based assessment of the partial factor for shear design of fibre reinforced concrete members without shear reinforcement

Fibre reinforced concrete (FRC) is increasingly used for structural purposes owing to its many benefits, especially in terms of improved overall sustainability of FRC structures relative to traditional reinforced concrete (RC). Such increased structural use of FRC requires safe and reliable models for its design in ultimate limit states (ULS). Particularly important are models for shear strength of FRC members without shear resistance due to the potential of brittle failure. The fib Model Code 2010 contains a model for the shear strength of FRC members without shear reinforcement and the same partial factor accepted for RC structures is accepted for FRC elements. This approach, however, is potentially on the unsafe side since the uncertainties of some design-determining mechanical properties of FRC (i.e., residual flexural strength) are larger than those for RC. Therefore, in this study, a comprehensive reliability-based calibration of the partial factor γc for the shear design of FRC members without shear reinforcement according to the fib Model Code 2010 model is performed. As a first step, the model error δ is assessed on 332 experimental results. Then, a parametric analysis of 700 cases is performed and a relationship between the target failure probability βR and γc is established. The results demonstrate that the current model together with the prescribed value of γc = 1.50 does not comply with the failure probabilities accepted for the different consequences of failure of FRC members over a 50-year service life. Therefore, changes to the shear resistance model are proposed in order to achieve the target failure probabilities.

Nonstationary Extreme Value Analysis of Nearshore Sea-State Parameters under the Effects of Climate Change: Application to the Greek Coastal Zone and Port Structures

In the present work, a methodological framework, based on nonstationary extreme value analysis of nearshore sea-state parameters, is proposed for the identification of climate change impacts on coastal zone and port defense structures. The applications refer to the estimation of coastal hazards on characteristic Mediterranean microtidal littoral zones and the calculation of failure probabilities of typical rubble mound breakwaters in Greek ports. The proposed methodology hinges on the extraction of extreme wave characteristics and sea levels due to storm events affecting the coast, a nonstationary extreme value analysis of sea-state parameters and coastal responses using moving time windows, a fitting of parametric trends to nonstationary parameter estimates of the extreme value models, and an assessment of nonstationary failure probabilities on engineered port protection. The analysis includes estimation of extreme total water level (TWL) on several Greek coasts to approximate the projected coastal flooding hazard under climate change conditions in the 21st century. The TWL calculation considers the wave characteristics, sea level height due to storm surges, mean sea level (MSL) rise, and astronomical tidal ranges of the study areas. Moreover, the failure probabilities of a typical coastal defense structure are assessed for several failure mechanisms, considering variations in MSL, extreme wave climates, and storm surges in the vicinity of ports, within the framework of reliability analysis based on the nonstationary generalized extreme value (GEV) distribution. The methodology supports the investigation of future safety levels and possible periods of increased vulnerability of the studied structure to different ultimate limit states under extreme marine weather conditions associated with climate change, aiming at the development of appropriate upgrading solutions. The analysis suggests that the assumption of stationarity might underestimate the total failure probability of coastal structures under future extreme marine conditions.

Benchmarking Study on Probabilistic Fracture Mechanics Analysis Codes xLPR and PASCAL-SP Considering Primary Water Stress Corrosion Cracking

Probabilistic fracture mechanics (PFM) is expected as a more rational methodology for the structural integrity assessments of nuclear power components because it can consider the inherent probabilistic distributions of various influencing factors and quantitatively evaluate the failure probabilities of the components. The Japan Atomic Energy Agency (JAEA) has developed a PFM analysis code, PASCAL-SP, to evaluate the failure probabilities of piping caused by aging degradation mechanisms, such as fatigue and stress corrosion cracking in the environments of both pressurized water and boiling water reactors. To improve confidence in the analysis results obtained from PASCAL-SP, a benchmarking study was conducted together with the PFM analysis code, xLPR, which was developed jointly by the U.S. Nuclear Regulatory Commission (NRC) and the Electric Power Research Institute. The benchmarking study was composed of deterministic and probabilistic analyses related to primary water stress corrosion cracking in a dissimilar metal weld joint in a pressurized water reactor surge line. The analyses were conducted independently by NRC staff and JAEA using their own codes and under common analysis conditions. In the present paper, the analysis conditions for the deterministic and probabilistic analyses are described in detail, and the analysis results obtained from the xLPR and PASCAL-SP codes are presented. It was confirmed that the analysis results obtained from the two codes were in good agreement.

Some conditional reliability properties of k-out-of-n system composed of different types of components with discrete independent lifetimes

In this paper, we study reliability properties of k-out-of-n system consisting of l$$(1\le l\le n)$$ ( 1 ≤ l ≤ n ) different types of components with discrete, independent lifetimes. We obtain some conditional survival functions of lifetime of a used system. Next, we use them to calculate two conditional failure probabilities of k-out-of-n systems and show that they are equal to unconditional failure probability of a k-out-of-$$(n-r)$$ ( n - r ) system, $$r<n-k+1$$ r < n - k + 1 . These results are extended versions of the respective ones existing in the literature.

Digital Scanning of Welds and Influence of Sampling Resolution on the Predicted Fatigue Performance: Modelling, Experiment and Simulation

Digital weld quality assurance systems are increasingly used to capture local geometrical variations that can be detrimental for the fatigue strength of welded components. In this study, a method is proposed to determine the required scanning sampling resolution for proper fatigue assessment. Based on FE analysis of laser-scanned welded joints, fatigue failure probabilities are computed using a Weakest-link fatigue model with experimentally determined parameters. By down-sampling of the scanning data in the FE simulations, it is shown that the uncertainty and error in the fatigue failure probability prediction increases with decreased sampling resolution. The required sampling resolution is thereafter determined by setting an allowable error in the predicted failure probability. A sampling resolution of 200 to 250 μm has been shown to be adequate for the fatigue-loaded welded joints investigated in the current study. The resolution requirements can be directly incorporated in production for continuous quality assurance of welded structures. The proposed probabilistic model used to derive the resolution requirement accurately captures the experimental fatigue strength distribution, with a correlation coefficient of 0.9 between model and experimental failure probabilities. This work therefore brings novelty by deriving sampling resolution requirements based on the influence of stochastic topographical variations on the fatigue strength distribution.