Investigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels

Resistance spot welding is an established joining process for the production of safety-relevant components in the automotive industry. Therefore, consecutive process monitoring is essential to meet the high quality requirements. Artificial neural networks can be used to evaluate the process paramete...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Bassel El-Sari, Max Biegler, Michael Rethmeier
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	automotive resistance spot welding quality assurance quality monitoring artificial intelligence Mining engineering. Metallurgy TN1-997
Acceso en línea:	https://doaj.org/article/789cb2fc227c460bb59592921b2aa342
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:789cb2fc227c460bb59592921b2aa342
record_format	dspace
spelling	oai:doaj.org-article:789cb2fc227c460bb59592921b2aa3422021-11-25T18:22:34ZInvestigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels10.3390/met111118742075-4701https://doaj.org/article/789cb2fc227c460bb59592921b2aa3422021-11-01T00:00:00Zhttps://www.mdpi.com/2075-4701/11/11/1874https://doaj.org/toc/2075-4701Resistance spot welding is an established joining process for the production of safety-relevant components in the automotive industry. Therefore, consecutive process monitoring is essential to meet the high quality requirements. Artificial neural networks can be used to evaluate the process parameters and signals, to ensure individual spot weld quality. The predictive accuracy of such algorithms depends on the provided training data set, and the prediction of untrained data is challenging. The aim of this paper was to investigate the extrapolation capability of a multi-layer perceptron model. That means, the predictive performance of the model was tested with data that clearly differed from the training data in terms of material and coating composition. Therefore, three multi-layer perceptron regression models were implemented to predict the nugget diameter from process data. The three models were able to predict the training datasets very well. The models, which were provided with features from the dynamic resistance curve predicted the new dataset better than the model with only process parameters. This study shows the beneficial influence of process signals on the predictive accuracy and robustness of artificial neural network algorithms. Especially, when predicting a data set from outside of the training space.Bassel El-SariMax BieglerMichael RethmeierMDPI AGarticleautomotiveresistance spot weldingquality assurancequality monitoringartificial intelligenceMining engineering. MetallurgyTN1-997ENMetals, Vol 11, Iss 1874, p 1874 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	automotive resistance spot welding quality assurance quality monitoring artificial intelligence Mining engineering. Metallurgy TN1-997
spellingShingle	automotive resistance spot welding quality assurance quality monitoring artificial intelligence Mining engineering. Metallurgy TN1-997 Bassel El-Sari Max Biegler Michael Rethmeier Investigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels
description	Resistance spot welding is an established joining process for the production of safety-relevant components in the automotive industry. Therefore, consecutive process monitoring is essential to meet the high quality requirements. Artificial neural networks can be used to evaluate the process parameters and signals, to ensure individual spot weld quality. The predictive accuracy of such algorithms depends on the provided training data set, and the prediction of untrained data is challenging. The aim of this paper was to investigate the extrapolation capability of a multi-layer perceptron model. That means, the predictive performance of the model was tested with data that clearly differed from the training data in terms of material and coating composition. Therefore, three multi-layer perceptron regression models were implemented to predict the nugget diameter from process data. The three models were able to predict the training datasets very well. The models, which were provided with features from the dynamic resistance curve predicted the new dataset better than the model with only process parameters. This study shows the beneficial influence of process signals on the predictive accuracy and robustness of artificial neural network algorithms. Especially, when predicting a data set from outside of the training space.
format	article
author	Bassel El-Sari Max Biegler Michael Rethmeier
author_facet	Bassel El-Sari Max Biegler Michael Rethmeier
author_sort	Bassel El-Sari
title	Investigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels
title_short	Investigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels
title_full	Investigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels
title_fullStr	Investigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels
title_full_unstemmed	Investigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels
title_sort	investigation of the extrapolation capability of an artificial neural network algorithm in combination with process signals in resistance spot welding of advanced high-strength steels
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/789cb2fc227c460bb59592921b2aa342
work_keys_str_mv	AT basselelsari investigationoftheextrapolationcapabilityofanartificialneuralnetworkalgorithmincombinationwithprocesssignalsinresistancespotweldingofadvancedhighstrengthsteels AT maxbiegler investigationoftheextrapolationcapabilityofanartificialneuralnetworkalgorithmincombinationwithprocesssignalsinresistancespotweldingofadvancedhighstrengthsteels AT michaelrethmeier investigationoftheextrapolationcapabilityofanartificialneuralnetworkalgorithmincombinationwithprocesssignalsinresistancespotweldingofadvancedhighstrengthsteels
_version_	1718411267107979264

Investigation of the Extrapolation Capability of an Artificial Neural Network Algorithm in Combination with Process Signals in Resistance Spot Welding of Advanced High-Strength Steels

Ejemplares similares