Characterization of biocomposites and glass fiber epoxy composites based on acoustic emission signals, deep feature extraction, and machine learning

Kek, Tomaž (author), Potočnik, Primož (author), Misson, Martin (author), Bergant, Zoran (author), Sorgente, Mario (author), Govekar, Edvard (author), Šturm, Roman (author)

This study presents the results of acoustic emission (AE) measurements and characterization in the loading of biocomposites at room and low temperatures that can be observed in the aviation industry. The fiber optic sensors (FOS) that can outperform electrical sensors in challenging operational environments were used. Standard features were extracted from AE measurements, and a convolutional autoencoder (CAE) was applied to extract deep features from AE signals. Different machine learning methods including discriminant analysis (DA), neural networks (NN), and extreme learning machines (ELM) were used for the construction of classifiers. The analysis is focused on the classification of extracted AE features to classify the source material, to evaluate the predictive importance of extracted features, and to evaluate the ability of used FOS for the evaluation of material behavior under challenging low-temperature environments. The results show the robustness of different CAE configurations for deep feature extraction. The combination of classic and deep features always significantly improves classification accuracy. The best classification accuracy (80.9%) was achieved with a neural network model and generally, more complex nonlinear models (NN, ELM) outperform simple models (DA). In all the considered models, the selected combined features always contain both classic and deep features.

• Language: English
• Keywords: polymer composites, biocomposites, GFE composites, acoustic emission, deep feature extraction, convolutional autoencoder, machine learning, neural networks
• Work type: Article
• Typology: 1.01 - Original Scientific Article
• Organization: FS - Faculty of Mechanical Engineering
• Publication status: Published
• Publication version: Version of Record
• Publication date: 13.09.2022
• Year: 2022
• Number of pages: 16 str.
• Numbering: Vol. 22, iss. 18, art. 6886
• PID: 20.500.12556/RUL-140622
• UDC: 620.168:620.179.17:007.52
• ISSN on article: 1424-8220
• DOI: 10.3390/s22186886
• COBISS.SI-ID: 121597443
• Publication date in RUL: 16.09.2022

Record is a part of a journal

• Title: Sensors
• Shortened title: Sensors
• Publisher: MDPI
• ISSN: 1424-8220
• COBISS.SI-ID: 10176278

Licences

• License: CC BY 4.0, Creative Commons Attribution 4.0 International
• Link: http://creativecommons.org/licenses/by/4.0/
• Description: This is the standard Creative Commons license that gives others maximum freedom to do what they want with the work as long as they credit the author.
• Licensing start date: 13.09.2022

Secondary language

• Language: Slovenian
• Title: Izpeljava globokih značilk na osnovi signalov AE za karakterizacijo obremenjenih epoksidnih kompozitov iz ogljikovih vlaken in epoksidnih kompozitov iz steklenih vlaken.

Abstract

Ta študija predstavlja rezultate meritev akustične emisije (AE) in karakterizacijo pri obremenjevanju biokompozitov pri sobnih in nizkih temperaturah, ki jih lahko opazimo v letalski industriji. Uporabljeni so bili senzorji z optičnimi vlakni (FOS), ki lahko prekašajo električne senzorje v zahtevnih delovnih okoljih. Standardne značilke so bile pridobljene iz meritev AE, za pridobivanje globokih značilk iz signalov AE pa je bil uporabljen konvolucijski autoenkoder (CAE). Za izdelavo klasifikatorjev so bile uporabljene različne metode strojnega učenja, vključno z diskriminantno analizo (DA), nevronskimi mrežami (NN) in ekstremnimi učnimi stroji (ELM). Analiza je osredotočena na klasifikacijo izpeljanih AE značilk za razvrstitev izvornega materiala, za ovrednotenje prediktivne pomembnosti izpeljanih značilk in za ovrednotenje zmožnosti uporabljenega FOS za oceno obnašanja materiala v zahtevnih nizkotemperaturnih okoljih. Rezultati kažejo robustnost različnih konfiguracij CAE za izpeljavo globokih zanačilk. Kombinacija klasičnih in globokih značilk vedno bistveno izboljša natančnost klasifikacije. Najboljša klasifikacijska natančnost (80,9 %) je bila dosežena z modelom nevronske mreže in na splošno so kompleksnejši nelinearni modeli (NN, ELM) boljši od enostavnih modelov (DA). Pri vseh obravnavanih modelih izbrane kombinirane značilke vedno vsebujejo tako klasične kot tudi globoke značilke.

• Keywords: polimerni kompoziti, biokompoziti, GFE kompoziti, akustična emisija, izpeljava globokih značilk, konvolucijski autoenkoder, strojno učenje, nevronske mreže

Projects

• Funder: ARRS - Slovenian Research Agency
• Project number: P2-0241
• Name: Sinergetika kompleksnih sistemov in procesov

• Funder: ARRS - Slovenian Research Agency
• Project number: P2-0270
• Name: Proizvodni sistemi, laserske tehnologije in spajanje materialov