Data-based structural health monitoring using small training data sets

2015 ◽  
Vol 22 (10) ◽  
pp. 1240-1264 ◽  
Author(s):  
Luciana Balsamo ◽  
Raimondo Betti
Keyword(s):  
Structural Health Monitoring ◽  
Health Monitoring ◽  
Training Data ◽  
Data Sets ◽  
Structural Health

scholarly journals Load-Identification Method for Flexible Multiple Corrugated Skin Using Spectra Features of FBGs

Aerospace ◽  
2021 ◽  
Vol 8 (5) ◽  
pp. 134
Author(s):  
Zhaoyu Zheng ◽  
Jiyun Lu ◽  
Dakai Liang
Keyword(s):  
Structural Health Monitoring ◽  
Fiber Bragg Grating ◽  
Health Monitoring ◽  
Training Data ◽  
Bragg Grating ◽  
Load Identification ◽  
Structural Health ◽  
Fiber Bragg Grating Sensors ◽  
Identification Method ◽  
Load Size

Flexible corrugated skins are ideal structures for morphing wings, and the associated load measurements are of great significance in structural health monitoring. This paper proposes a novel load-identification method for flexible corrugated skins based on improved Fisher discrimination dictionary learning (FDDL). Several fiber Bragg grating sensors are pasted on the skin to monitor the load on multiple corrugated crests. The loads on different crests cause nonuniform strain fields, and these discriminative spectra are recorded and used as training data. The proposed method involves load-positioning and load-size identification. In the load-size-identification stage, a classifier is trained for every corrugated crest. An interleaved block grouping of samples is introduced to enhance the discrimination of dictionaries, and a two-resolution load-size classifier is introduced to improve the performance and resolution of the grouping labels. An adjustable weight is introduced to the FDDL classification scheme to optimize the contribution from different sensors for different load-size classifiers. With the proposed method, the individual loads on eight crests can be identified by two fiber Bragg grating sensors. The positioning accuracy is 100%, and the mean error of the load-size identification is 0.2106 N, which is sufficiently precise for structural health monitoring.

scholarly journals Structural health monitoring algorithm comparisons using standard data sets

2009 ◽  
Author(s):  
Eloi Figueiredo ◽  
Gyuhae Park ◽  
Joaquim Figueiras ◽  
Charles Farrar ◽  
Keith Worden
Keyword(s):  
Structural Health Monitoring ◽  
Health Monitoring ◽  
Data Sets ◽  
Structural Health ◽  
Standard Data ◽  
Monitoring Algorithm

Structural Health Monitoring of Composite Structures via Machine Learning of Mechanoluminescence

2019 ◽  
Author(s):  
Sujasha Gupta ◽  
Srivatsava Krishnan ◽  
Vishnubaba Sundaresan
Keyword(s):  
Machine Learning ◽  
Elastic Modulus ◽  
Structural Health Monitoring ◽  
Health Monitoring ◽  
Learning Algorithm ◽  
Structural Response ◽  
Training Data ◽  
Machine Learning Algorithm ◽  
Distributed Sensors ◽  
Structural Health

Abstract The goal of this paper is to develop a machine learning algorithm for structural health monitoring of polymer composites with mechanoluminescent phosphors as distributed sensors. Mechanoluminescence is the phenomenon of light emission from organic/inorganic materials due to mechanical stimuli. Distributed sensors collect a large amount of data and contain structural response information that is difficult to analyze using classical or continuum models. Hence, approaches to analyze this data using machine learning or deep learning is necessary to develop models that describe initiation of damage, propagation and ultimately structural failure. This paper focuses on developing a machine learning algorithm that predicts the elastic modulus of a structure as a function of input parameters such as stress and measured light output. The training data for the algorithm utilizes experimental results from cyclical loading of elastomeric composite coupons impregnated with ML particles. A multivariate linear regression is performed on the elastic modulus within the training data as a function of stress and ML emission intensity. Error in predicted elastic modulus is minimized using a gradient descent algorithm. The machine learning algorithm outlined in this paper is expected to provide insights into structural response and deterioration of mechanical properties in real-time that cannot be obtained using a finite array of sensors.

Efficient storage and processing of large guided wave data sets with random projections

2020 ◽  
pp. 147592172096019
Author(s):  
Sungwon Kim ◽  
Spencer Shiveley ◽  
Alexander CS Douglass ◽  
Yisong Zhang ◽  
Rajeev Sahay ◽  
...  
Keyword(s):  
Structural Health Monitoring ◽  
Damage Detection ◽  
Health Monitoring ◽  
Random Projection ◽  
Guided Wave ◽  
Random Projections ◽  
Data Sets ◽  
Structural Health ◽  
Efficient Storage ◽  
Value Decomposition

Over the last several decades, structural health monitoring systems have grown into increasingly diverse applications. Structural health monitoring excels with large data sets that can capture the typical variability, novel events, and undesired degradation over time. As a result, the efficient storage and processing of these large, guided wave data sets have become a key feature for successful application of structural health monitoring. This article describes a series of investigations into the use of random projection theory to significantly reduce storage burdens and improve computational complexity while not significantly affecting common damage detection strategies. Random projections are used as a lossy compression scheme that approximately retains metrics of distance or similarity between data records. Random projection compression is evaluated using a large 1,440,000 measurement data set, which was collected over 5 months in an unprotected outdoor environment. Accurate damage detection, after the compression process, is achieved through correlation analysis and singular value decomposition. The results indicate consistent detection performance with over 95% of storage compression and more than a 477 times speed improvement in computational cost for singular value decomposition–based damage detection.

Sign in / Sign up

Export Citation Format

Share Document