A particle filter-based model selection algorithm for fatigue damage identification on aeronautical structures

Cadini, Francesco; Sbarufatti, Claudio; Corbetta, Matteo; Giglio, Marco

doi:10.1002/stc.2002

Cited by 21 publications

(22 citation statements)

References 21 publications

(63 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In general, the availability of accurate physics‐based, or empirical/phenomenological, models accounting for several environmental and operating conditions and, possibly, tailored to the specific structural degradation problem under analysis is a fundamental advantage with respect to the use of purely data‐driven methods, such as those based on machine learning techniques. This is thoroughly demonstrated by quite a large amount of literature, often resorting to Bayesian frameworks (e.g., previous studies), and the relative performances are even compared in some works, such as Baraldi et al . However, as anticipated above, problems may arise when one‐of‐a‐kind applications are dealt with, for which no analytical models are available, or when the degradation behavior of an individual component of a family of kin is significantly different from expected, due to the operating/environmental conditions or variability of manufacturing processes.…”

Section: Introductionmentioning

confidence: 95%

Particle filtering‐based adaptive training of neural networks for real‐time structural damage diagnosis and prognosis

Cadini

Sbarufatti

Corbetta

et al. 2019

Struct Control Health Monit

Self Cite

View full text Add to dashboard Cite

Real-time failure diagnosis and prognosis play key roles in prognostics and health management of critical engineering assets. Physics-based (or empiricalphenomenological) and data-driven methods have been proposed in the literature to solve damage diagnosis and prognosis challenges. Among data-driven methods, neural networks have often successfully proven their regression capabilities for the complex, nonlinear problems typical of this field research field. Indeed, the need to resort to machine learning techniques (e.g., neural networks) becomes stronger when physics-based (or empirical-phenomenological) models are lacking for describing the degradation behavior of the systems under consideration. In this work, we propose to exploit the flexibility offered by neural networks in order to adaptively, and in real time, learn from the monitored structure and derive models for diagnosis and prognosis of structural components subject to damage processes. To this aim, neural networks are embedded within a particle filtering scheme, and the training of the network is performed sequentially and in real time, as observations become available during the damage evolution. The structural health monitoring tool is capable of sequentially updating itself using the observations gathered during the system's operation, avoiding the typical limitations associated to off-line training schemes, which are, in general, not capable of capturing possible changes in the dynamical degradation behavior, when not properly included in the training patterns. For validation and comparison purposes, the method is demonstrated on simulated fatigue crack growths and real crack evolution dynamics observed in a metallic aeronautical panel loaded with a changing-amplitude fatigue load, during an accelerated laboratory test. KEYWORDSadaptive diagnostic and prognostic, neural networks, particle filtering, remaining useful life Abbreviations FCG fatigue crack growth MLP multilayer perceptron NN neural network pdf probability density function PF particle filtering RUL remaining useful life | INTRODUCTIONIn the last decades, modern industry has fostered a significant increase in the availability and reliability of components and structures in order to achieve more robust financial sustainability and, at the same time, higher safety levels. In this context, real-time failure diagnosis and prognosis play key roles in prognostics and health management of critical engineering assets, where the forecast of the system's degradation drives maintenance decision making and instantaneous safety assessments. In order to perform these tasks, a large variety of both model-based (either physical, empirical, or phenomenological) and data-driven methods have been proposed in the literature.In general, the availability of accurate physics-based, or empirical/phenomenological, models accounting for several environmental and operating conditions and, possibly, tailored to the specific structural degradation problem under analysis is a fundamental advantage with respect to t...

show abstract

Section: Introductionmentioning

confidence: 95%

Particle filtering‐based adaptive training of neural networks for real‐time structural damage diagnosis and prognosis

Cadini

Sbarufatti

Corbetta

et al. 2019

Struct Control Health Monit

Self Cite

View full text Add to dashboard Cite

show abstract

“…Yin et al [17] applied a state resonance algorithm-based stochastic resonance method to high-speed train traction systems to achieve optimal extraction and frequency recovery. Cadini et al [18] used a particle filter to estimate the length of cracks to diagnose the early faults of aeronautical structure cracks. For state estimation with unknown inputs, some researchers use unknown input residual generators and estimation filters to achieve fault diagnosis [19].…”

Section: Introductionmentioning

confidence: 99%

State-Degradation-Oriented Fault Diagnosis for High-Speed Train Running Gears System

Cheng

Wang

Luo

et al. 2020

Sensors

View full text Add to dashboard Cite

As one of the critical components of high-speed trains, the running gears system directly affects the operation performance of the train. This paper proposes a state-degradation-oriented method for fault diagnosis of an actual running gears system based on the Wiener state degradation process and multi-sensor filtering. First of all, for the given measurements of the high-speed train, this paper considers the information acquisition and transfer characteristics of composite sensors, which establish a distributed topology for axle box bearing. Secondly, a distributed filtering is built based on the bilinear system model, and the gain parameters of the filter are designed to minimize the mean square error. For a better presentation of the degradation characteristics in actual operation, this paper constructs an improved nonlinear model. Finally, threshold is determined based on the Chebyshev’s inequality for a reliable fault diagnosis. Open datasets of rotating machinery bearings and the real measurements are utilized in the case studies to demonstrate the effectiveness of the proposed method. Results obtained in this paper are consistent with the actual situation, which validate the proposed methods.

show abstract

“…18 In addition to these methods, hierarchical temporal memory mimicking the architecture and processes of cortical neurons shows potential for the real-time anomaly detection. 19 In the field of civil engineering, several studies have adopted machine learning methods such as particle filter-based model, 20,21 extended Kalman filter-based model, 22 and Bayesian probabilistic approach 23 for online learning purpose. Existing applications of such methods typically require specific information about the structure that is not suited for a widespread deployment across thousands of bridges and dams that are all different from one to another.…”

Section: Introductionmentioning

confidence: 99%

Real‐time anomaly detection with Bayesian dynamic linear models

Nguyen

Goulet

2019

Struct Control Health Monit

View full text Add to dashboard Cite

Summary A key goal in structural health monitoring is to detect abnormal events in a structure's behavior by interpreting its observed responses over time. The goal is to develop an anomaly detection method that (a) is robust towards false alarm and (b) capable of performing real‐time analysis. The majority of anomaly detection approaches are currently operating over batches of data for which the model parameters are assumed to be constant over time and to be equal to the values estimated during a fixed‐size training period. This assumption is not suited for the real‐time anomaly detection where model parameters need to be treated as time‐varying quantities. This paper presents how this issue is tackled by combining Rao‐Blackwellized particle filter (RBPF) with the Bayesian dynamic linear models (BDLMs). The BDLMs, which is a special case of state‐space models, allow decomposing time series into a vector of hidden state variables. The RBPF employs the sequential Monte Carlo method to learn model parameters continuously as the new observations are collected. The potential of the new approach is illustrated on the displacement data collected from a dam in Canada. The approach succeeds in detecting the anomaly caused by the refection work on the dam as well as the artificial anomalies that are introduced on the original dataset. The new method opens the way for monitoring the structure's health and conditions in real time.

show abstract

A particle filter-based model selection algorithm for fatigue damage identification on aeronautical structures

Cited by 21 publications

References 21 publications

Particle filtering‐based adaptive training of neural networks for real‐time structural damage diagnosis and prognosis

Particle filtering‐based adaptive training of neural networks for real‐time structural damage diagnosis and prognosis

State-Degradation-Oriented Fault Diagnosis for High-Speed Train Running Gears System

Real‐time anomaly detection with Bayesian dynamic linear models

Contact Info

Product

Resources

About