Machine learning assisted mechanical design: Multi objective optimization of welding parameters for robotic gripper joint

Basit Öğe Kaydı
dc.authorid0000-0002-1723-4108
dc.contributor.authorTanrıver, Kürşat
dc.date.accessioned2026-04-15T10:13:48Z
dc.date.available2026-04-15T10:13:48Z
dc.date.issued2026
dc.departmentFakülteler, Mühendislik ve Doğa Bilimleri Fakültesi, Mekatronik Mühendisliği Bölümü
dc.description.abstractThis study presents a data driven, mechanically grounded optimization framework to improve the tensile performance of TIG welded AISI 304 stainless steel joints used in load bearing robotic gripper applications. Unlike prior machine learning–assisted welding optimization studies that emphasize metallurgical characteristics or isolated statistical metrics, the proposed approach explicitly links welding parameters to structural per formance and mechanical design requirements of robotic systems. A Taguchi based experimental design consisting of ten orthogonally arranged welding trials was employed to generate the training dataset. Gaussian Process Regression (GPR) and Support Vector Regression (SVR) models were developed to predict ultimate tensile strength (UTS) and eval uated using cross validation metrics. The GPR model demonstrated super ior accuracy, achieving a coefficient of determination exceeding 0.97 with low prediction error. Feature importance analysis identified welding volt age and material thickness as dominant parameters, consistent with their influence on arc stability, heat distribution, weld pool geometry and load bearing cross sectional integrity. Multi objective optimization was performed using a Multi Objective Genetic Algorithm (MOGA), where UTS was treated as the primary mechan ical objective, while heat input and a distortion related surrogate response were included as competing process related objectives. The resulting Pareto optimal solutions yielded UTS values of approximately 554–580 MPa within feasible parameter limits. Model robustness was assessed through Monte Carlo–based uncertainty quantification using ±5% stochastic variation in welding parameters. Two confirmation experiments conducted outside the training dataset demon strated agreement between predicted and experimental UTS values, with total prediction errors below 1.3%. Overall, the proposed ML-MOGA frame work offers a practical optimization strategy for reliable decision making under limited experimental data conditions.
dc.identifier.citationTanrıver, K. (2026). Machine learning assisted mechanical design: Multi objective optimization of welding parameters for robotic gripper joint. Mechanics Based Design of Structures and Machines, 54(1), pp. 1-30. https://doi.org/10.1080/15397734.2026.2643458
dc.identifier.doi10.1080/15397734.2026.2643458
dc.identifier.endpage30
dc.identifier.issn1539-7734
dc.identifier.issn1539-7742
dc.identifier.issue1
dc.identifier.scopus2-s2.0-105033436108
dc.identifier.scopusqualityQ1
dc.identifier.startpage1
dc.identifier.urihttps://doi.org/10.1080/15397734.2026.2643458
dc.identifier.urihttps://hdl.handle.net/20.500.13055/1410
dc.identifier.volume54
dc.identifier.wosWOS:001719239100001
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.indekslendigikaynak.otherSCI-E - Science Citation Index Expanded
dc.institutionauthorTanrıver, Kürşat
dc.institutionauthorid0000-0002-1723-4108
dc.language.isoen
dc.publisherTaylor & Francis
dc.relation.ispartofMechanics Based Design of Structures and Machines
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.subjectMachine Learning
dc.subjectOptimization
dc.subjectTaguchi Method
dc.subjectRobot
dc.subjectTIG Welding
dc.titleMachine learning assisted mechanical design: Multi objective optimization of welding parameters for robotic gripper joint
dc.typeArticle
dspace.entity.typePublication

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Kapalı Erişim
İsim:
Tam Metin / Full Text.pdf
Boyut:
2.86 MB
Biçim:
Adobe Portable Document Format
İndir
Lisans paketi
Listeleniyor 1 - 1 / 1
Kapalı Erişim
İsim:
license.txt
Boyut:
1.17 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
İndir

Koleksiyon

Mekatronik Mühendisliği Bölümü Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu
WoS İndeksli Yayınlar Koleksiyonu