Novelty Detection Using Sparse Auto-Encoders to Characterize Structural Vibration Responses

Finotti, Rafaelle Piazzaroli; Barbosa, Fernando Policarpo; Cury, Alexandre; Pimentel, Roberto Leal

doi:10.1007/s13369-022-06732-6

Cited by 5 publications

(2 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, the random load-time history of the catenary is loaded with the lifting force at different speeds and weight levels obtained from the dynamic simulation of the railway wagon, and the loading position is the portal structure beam. The dynamic load generated during the operation of the railway wagon and the load that the railway wagon will generate during driving can be composed of excitation function, static load and railway wagon vibration series functions [25,26]:…”

Section: Dynamic Load Loading Analysismentioning

confidence: 99%

Vibration Characteristics Analysis and Structure Optimization of Catenary Portal Structure on Four-Wire Bridge

Wang¹,

Li²

2022

Structural Durability &Amp; Health Monitoring

View full text Add to dashboard Cite

The portal structure is the support equipment in the catenary, which bears the load of contact suspension and support equipment. In practical work, with the change of external forces, the support equipment bears complex and changeable loads, so it has higher requirements for its reliability and safety. In order to study the dynamic characteristics of catenary portal structure on continuous beam of four-way bridge, taking the catenary portal structure on Dshaping four-way bridge as the research object, the portal structure simulation model of bridgenetwork integration was established in Midas Civil. The maximum point of deformation and stress was determined by finite element analysis of catenary hard span equipment, and the frequency and mode of natural vibration of hard span were obtained by modal analysis. Secondly, through the field dynamic stress acquisition test, combined with the results of finite element analysis, the fault location is determined, and the vibration characteristics are analyzed. Finally, based on the results of modal analysis and vibration analysis, the method that the vibration of portal structure beam is affected by structural stiffness and vibration frequency amplitude is proposed. The torsional vibration of the portal structure beam was suppressed by increasing the stiffness of the beam and reducing the vibration conduction between the trolley and the beam, and the hard cross beam was optimized by strengthening the hanging column and the connecting beam and adding diagonal support between the pillar and the portal structure beam. By comparing the values of shear, bending moment, displacement and dynamic stress on the hard span before and after optimization, the amplitude peak after structural optimization is reduced by about 25%, and the application of oblique support and reinforcement of the beam can significantly improve the portal structure vibration.

show abstract

Section: Dynamic Load Loading Analysismentioning

confidence: 99%

Vibration Characteristics Analysis and Structure Optimization of Catenary Portal Structure on Four-Wire Bridge

Wang¹,

Li²

2022

Structural Durability &Amp; Health Monitoring

View full text Add to dashboard Cite

show abstract

“…Do et al [18] demonstrated the capability of autoencoders based on a long short-term memory structure for detecting anomaly vibrations in a Vertical Carousel Storage and Retrieval System for industrial applications. Finotti et al [19] assessed the structural condition of a viaduct by means of a sparse autoencoder that learned important data features (to characterize the vibration signals) and a support vector machine that classified the corresponding damage based on the extracted features. Jimenez-Martinez et al improved fatigue life prediction through a combination of synthetic data and an ANN without requiring additional tests or new parameters by overcoming the limit of Miner's damage rule when taking into account different factors such as temperature, environment, sequence loads, and mean stress [20].…”

Section: Introductionmentioning

confidence: 99%

A Comparison of Deep Learning Algorithms for Anomaly Detection in Discrete Mechanical Systems

et al. 2023

View full text Add to dashboard Cite

The application of intelligent systems for structural health monitoring is investigated. A change in the nominal configuration can be related to a structural defect that has to be monitored before it reaches a critical condition. Evidently, the ability to automatically detect changes in a structure is a very attractive feature. When there is no prior knowledge on the system, deep learning models could effectively detect a change and enhance the capability of determining the damage location. However, the acquisition of data related to damaged structures is not always practical. In this paper, two deep learning approaches, a physics-informed autoencoder and a simple data-driven autoencoder, are applied to a test rig consisting of a small four-storey building model. Modifications to the system are simulated by changing the stiffness of the springs. Both the machine learning algorithms outperform the traditional approach based on an experimental modal analysis. Moreover, the increased potential of the physics-informed neural networks to detect and locate damage is confirmed.

show abstract

A Statistical Approach to Hyperparameter Tuning of Deep Learning for Construction Machine Classification

Ottoni,

Novo,

Oliveira

2023

Arab J Sci Eng

View full text Add to dashboard Cite

Novelty Detection Using Sparse Auto-Encoders to Characterize Structural Vibration Responses

Cited by 5 publications

References 39 publications

Vibration Characteristics Analysis and Structure Optimization of Catenary Portal Structure on Four-Wire Bridge

Vibration Characteristics Analysis and Structure Optimization of Catenary Portal Structure on Four-Wire Bridge

A Comparison of Deep Learning Algorithms for Anomaly Detection in Discrete Mechanical Systems

A Statistical Approach to Hyperparameter Tuning of Deep Learning for Construction Machine Classification

Contact Info

Product

Resources

About