Lazer Kesimde Kesme Genişliğini Tahmin Etmek için Bilgi Tabanlı Bulanık Mantık Yönteminin Veri Odaklı Makine Öğrenmesi Modelleriyle Karşılaştırılması

Öz

Bu çalışma, nispeten küçük veri kısıtlamaları altında, veri odaklı yaygın makine öğrenimi modellerinin genelleme yeteneklerini, bilgi tabanlı bulanık mantık sisteminin yetenekleriyle karşılaştırmayı amaçlamaktadır. 4 mm kalınlığında S235JR sac metal için deneysel bir veri seti elde edildi ve bu sette lazer kesim gücü, gaz basıncı ve odak uzaklığı için dört farklı seviyede kesim aralığı genişliği ölçüldü. Toplam 64 veri noktasıyla geleneksel bir ANOVA analizi gerçekleştirildi. Gaz basıncının kesim aralığı genişliğine en yüksek katkıyı (%38,71) sağladığı bulundu. Hassas lazer kesiminde kesim aralığı genişliğini tahmin etmek için, tüm modeller için aynı 50 veri kullanılarak, Kesim Aralığı Genişliği-Mamdani Bulanık Mantık (KW-MFL) modeli oluşturuldu ve veri odaklı makine öğrenimi yöntemleri olan SVR ve GPR eğitildi. Her modelin performansını titizlikle değerlendirmek için 14 farklı veri seti kullanılarak model doğrulaması yapıldı. Sonuçlara göre, KW-MFL modeli, kesme genişliği tahmininde RMSE-MAE değerlerimi SVR'ye kıyasla %27,95-%27,60 ve GPR'ye kıyasla %21,87-%19,05 oranında azalttı. Sonuçlar, veri odaklı modellerin veri yetersizliğiyle mücadele ettiğini ve tatmin edici performans sergileyemediğini ortaya koydu. KW-MFL, uzman kural tabanına entegre olarak belirsizlikleri başarıyla telafi etti ve en düşük tahmin hatalarıyla en yüksek temsil yeteneğine ulaştı.

Anahtar Kelimeler

• Lazer kesim
• Kesme Genişliği
• Tahmin
• SVR
• GPR
• Mamdani Bulanık Mantık

Teşekkür

Yazarlar, bu çalışmanın deneysel aşaması için gerekli olan test numunelerini ve endüstriyel lazer kesim tesislerini sağladığı için MVD Machinery Company'ye teşekkür eder.

Benchmarking a Knowledge-Based Fuzzy with Data-Driven Machine Learning Models for Predicting Kerf Width in Laser Cutting

Öz

This study aims to compare the generalisation capabilities of data-driven common ML models with those of a knowledge-based fuzzy logic system under relatively small data constraints. An experimental dataset was obtained for a 4 mm-thick S235JR sheet metal in which the cutting kerf width was measured across four different levels for laser cutting power, gas pressure, and focal length. A conventional ANOVA analysis was performed with a total of 64 data points. The gas pressure was found to have the highest contribution to kerf width at 38.71%. To estimate the kerf width in precision laser cutting, using the same 50 data for all models, the Kerf Width-Mamdani Fuzzy Logic (KW-MFL) model was constructed, and SVR, GPR which are data-driven ML methods, were trained. Each model validation was performed using 14 unseen datasets to rigorously assess its performance. According to the results, the KW-MFL model reduced RMSE-MAE in kerf width estimation by 27.95%-27.60% to SVR and by 21.87%-19.05% to GPR, respectively. The results revealed that data-driven models struggled with data scarcity and not exhibit satisfactory performance. The KW-MFL successfully compensated for uncertainties by integrating into its expert rule base, achieving the highest representational ability with the lowest prediction errors.

Anahtar Kelimeler

• Laser cutting
• Kerf Width
• Prediction
• SVR
• GPR
• Mamdani Fuzzy Logic

Teşekkür

The authors thank MVD Machinery Company for providing the test specimens and industrial laser cutting facilities essential for the experimental phase of this study.

Kaynakça

1. [1] Syn Zhian C., Mokhtar M., Feng C.J., Manurung Y.H.P., “Approach to Prediction of Laser Cutting Quality by Employing Fuzzy Expert System”, Expert Systems with Applications, 38: 7558–7568, (2011).
2. [2] Rajaram N., Ahmad J.S., Cheraghi S.H., “CO2 Laser Cut Quality of 4130 Steel”, International Journal of Machine Tools and Manufacture, 43: 351–358, (2003).
3. [3] Çelebi A., “Deformation and microstructural analysis of fiber laser and TIG welding of thin Ti6Al4V sheet by coordinate measurement machine (CMM)”, Journal of Polytechnic, 23(4): 1183–1188, (2020).
4. [4] Madic M., Radovanovic M., Cojbasic Z., Nedic B., “Fuzzy Logic Approach for the Prediction of Dross Formation in CO2 Laser Cutting of Mild Steel”, Journal of Engineering Science and Technology Review, 8: 143–150, (2015).
5. [5] Madic M., Radovanovic M., Nedic B., “Modeling and Simulated Annealing Optimization of Surface Roughness in CO2 Laser Nitrogen Cutting of Stainless Steel”, Tribology in Industry, 35: 167–176, (2013).
6. [6] Dubey A.K., Yadava V., “Experimental Study of Nd:YAG Laser Beam Machining – An Overview”, Journal of Materials Processing Technology, 195: 15–26, (2008).
7. [7] Arif A.F.M., Yilbas B.S., Abdul Aleem B.J., “Laser Cutting of Thick Sheet Metals: Residual Stress Analysis”, Optics & Laser Technology, 41: 224–232, (2009).
8. [8] Fan J., Yuanyuan Z., Ri P., Jian Y., Xingfei R., Wanying Z., “Optimization of Laser Processing Parameters on the Width of Heat Affected Zone of Carbon Fiber Reinforced Polymer”, Journal of the Chinese Institute of Engineers, 45: 131–137, (2022).

9. [9] Joshi P., Sharma A., “Simultaneous Optimization of Kerf Taper and Heat Affected Zone in Nd-YAG Laser Cutting of Al 6061-T6 Sheet Using Hybrid Approach of Grey Relational Analysis and Fuzzy Logic”, Precision Engineering, 54: 302–313, (2018).
10. [10] Pandey A.K., Dubey A.K., “Taguchi Based Fuzzy Logic Optimization of Multiple Quality Characteristics in Laser Cutting of Duralumin Sheet”, Optics and Lasers in Engineering, 50: 328–335, (2012).
11. [11] Pandey A.K., Dubey A.K., “Multiple Quality Optimization in Laser Cutting of Difficult-To-Laser-Cut Material Using Grey-Fuzzy Methodology”, International Journal of Advanced Manufacturing Technology, 65: 421–431, (2013).
12. [12] Priyadarshini M., Tripathy P.P., Mishra D., Panda S., “Multi Characteristics Optimization of Laser Drilling Process Parameter Using Fuzzy-TOPSIS Method”, Materials Today: Proceedings, 4: 8538–8547, (2017).
13. [13] Tase Velazquez D.R., Helleno A.L., de Oliveira M.C., Fals H.C., Macias E.J., “Fuzzy Logic-Based Inference System for Prediction of Energy Input in Laser Metal Deposited AISI 316 Single-Beads”, 32nd European Modeling & Simulation Symposium (EMSS 2020), Almeria, Spain, 400–409, (2020).
14. [14] Pandey A.K., Dubey A.K., “Fuzzy Expert System for Prediction of Kerf Qualities in Pulsed Laser Cutting of Titanium Alloy Sheet”, Machining Science and Technology, 17: 545–574, (2013).
15. [15] Parthiban A., Sathish S., Ravikumar R., Prakash P., “Experimental Investigation and Fuzzy Logic Modelling of CO2 Laser Cutting Parameter for AA6061-T6 Sheet”, International Journal of Rapid Manufacturing, 9: 48–57, (2020).
16. [16] Chaudhary A.K., Pandey A.K., Dubey A.K., “Computer Aided Taguchi-Fuzzy Multi-Optimization of Laser Cutting Processes”, Journal of Intelligent and Fuzzy Systems, 26: 801–810, (2014).
17. [17] Sivarao P.B., Brevern P., El-Tayeb N.S.M., Vengkatesh V.C., “GUI Based Mamdani Fuzzy Inference System Modeling to Predict Surface Roughness in Laser Machining”, International Journal of Electrical & Computer Sciences, 9: 37–43, (2009).
18. [18] Patel P., Modi B.S., Sheth S., Patel T., “Experimental Investigation, Modelling and Comparison of Kerf Width in Laser Cutting of GFRP”, Bonfring International Journal of Industrial Engineering and Management Science, 5: 55–62, (2015).
19. [19] Rajamani D., Tamilarasan A., “Fuzzy and Regression Modelling for Nd:YAG Laser Cutting of Ti-6Al-4V Superalloy Sheet”, Journal for Manufacturing Science and Production, 16: 153–162, (2016).
20. [20] Petkovic D., Nikolic V., Milovancevic M., Lazov L., “Estimation of the Most Influential Factors on the Laser Cutting Process Heat Affected Zone (HAZ) by Adaptive Neuro-Fuzzy Technique”, Infrared Physics & Technology, 77: 12–15, (2016).
21. [21] Kusuma A.I., Huang Y.M., “Performance Comparison of Machine Learning Models for Kerf Width Prediction in Pulsed Laser Cutting”, International Journal of Advanced Manufacturing Technology, 123: 2703–2718, (2022).
22. [22] Vagheesan S., Govindarajulu J., “Comparative Regression and Neural Network Modeling of Roughness and Kerf Width in CO2 Laser Cutting of Aluminium”, Tehnički Vjesnik, 28: 1437–1441, (2021).
23. [23] Sridarane R., Kesavan S., Sankar R., Prakash P., Jagatheesan K., “Artificial Intelligence to Predict Kerf Width During CO2 Laser Cutting of Mild Steel”, Materials Today: Proceedings, 72: 2501–2506, (2023).
24. [24] Vasileska E., Tuteski O., Kusigerski B., Argilovski A., Tomov M., Gecevska V., “Statistical Analysis and Machine Learning-Based Modelling of Kerf Width in CO2 Laser Cutting of PMMA”, Manufacturing Technology, 24: 960–968, (2024).
25. [25] Özlü, B., “St-37 Malzemesinin Lazer Kesiminde İşleme Parametrelerinin Deneysel ve İstatiksel Olarak İncelenmesi”, Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 33(1):161-171, (2021).
26. [26] Cebeci, İ., Özlü, B., Demir, H. “AISI 304 paslanmaz sac malzemenin lazerle işlenmesinde kesme parametrelerinin yüzey pürüzlülüğü, kerf genişliği ve çapak yüksekliğine etkisinin araştırılması”, Niğde Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi, 11(2):410-415, (2022).
27. [27] Nasresfahani M.R., Asghari S., Ashani H.N., Gorabi H., Joharestani M.Z., “Exploring Alternative Manufacturing Techniques for High-Quality Solar Cell Interconnectors: A Comparative Study of Laser Cutting and Electric Discharge Machining”, Physica Scripta, 99: 121501, (2024).
28. [28] Parimanik S.R., Mishra D., Mahapatra T.R., Rathor S., Kant R., “NiTi Wire Welding Using Fe, Cu and Ni Fillers: Multi-Objective Modelling and AHP–Entropy Approach to Sustainability Evaluation”, Engineering Computations, 1–38, (2026).
29. [29] Hussain S., Qazi M.I., Abas M., “Investigation and Optimization of Plasma Arc Cutting Process Parameters for AISI 304 by Integrating Principal Component Analysis and Composite Desirability Method”, Journal of the Brazilian Society of Mechanical Sciences and Engineering, 46: 33, (2024).
30. [30] Joshi P., Sharma A., “Optimization of Process Parameters During Laser Beam Cutting of Ni-Based Superalloy Thin Sheet Along Curved Profile Using Grey-Fuzzy Methodology”, Journal of the Brazilian Society of Mechanical Sciences and Engineering, 40: 389, (2018).
31. [31] Oliveira M., Santos J.P., Almeida F.G., Reis A., Pereira J.P., Rocha A.B., “Impact of Laser-Based Technologies in the Energy-Consumption of Metal Cutters: Comparison Between Commercially Available Systems”, Key Engineering Materials, 473: 809–815, (2011).
32. [32] Soota, T., and Rajput S. K., "Optimization and measurement of kerf width and surface roughness of AISI 316L”, Forces in Mechanics, 6: 100071, (2022).
33. [33] Maity K.P., Bagal D.K., “Effect of Process Parameters on Cut Quality of Stainless Steel of Plasma Arc Cutting Using Hybrid Approach”, International Journal of Advanced Manufacturing Technology, 78: 161–175, (2015).
34. [34] Kumar J., Soota T., Rajput S.K., “Optimization and Measurement of Kerf Width and Surface Roughness of AISI 316L”, Forces in Mechanics, 6: 100071, (2022).
35. [35] Gupta V., Pandey P.M., Pal Garg M., Khanna R., Batra N.K., “Minimization of Kerf Taper Angle and Kerf Width Using Taguchi's Method in Abrasive Water Jet Machining of Marble”, Procedia Materials Science, 6: 140–149, (2014).
36. [36] Montesinos López O.A., Montesinos López A., Crossa J., “Support Vector Machines and Support Vector Regression”, Multivariate Statistical Machine Learning Methods for Genomic Prediction, Springer International Publishing, Cham, 337–378, (2022).
37. [37] Yadav A., Bareth R., Kochar M., Pazoki M., Sehiemy R.A.E., “Gaussian Process Regression-Based Load Forecasting Model”, IET Generation, Transmission & Distribution, 18: 899–910, (2024).
38. [38] Awad M., Khanna R., “Support Vector Regression”, Efficient Learning Machines: Theories, Concepts, and Applications for Engineers and System Designers, Apress, Berkeley, 67–80, (2015).
39. [39] Zhu Y., Meng F., Dai Y., Guo H., Shi C., Li Z., “Prediction of Cutting Surface Roughness of Compacted Graphite Iron (CGI) Based on Machine Learning Techniques”, Iranian Journal of Science and Technology, Transactions of Mechanical Engineering, 49: 1765–1772, (2025).
40. [40] Liu H., Ong Y.S., Shen X., Cai J., “When Gaussian Process Meets Big Data: A Review of Scalable GPs”, IEEE Transactions on Neural Networks and Learning Systems, 31: 4405–4423, (2020).
41. [41] Ouifak H., Idri A., “A Comprehensive Review of Fuzzy Logic Based Interpretability and Explainability of Machine Learning Techniques Across Domains”, Neurocomputing, 647: 130602, (2025).
42. [42] Öktem K., Tekerek A., “Bitcoin Price Direction Prediction Using Machine Learning on a Very Small Dataset”, Politeknik Dergisi, 28(6): 1731-1742, (2025).
43. [43] Zadeh L.A., “Outline of a New Approach to the Analysis of Complex Systems and Decision Processes”, IEEE Transactions on Systems, Man, and Cybernetics, 3: 28–44, (1973).
44. [44] Karkalos N.E., Galanis N.I., Markopoulos A.P., “Surface Roughness Prediction for the Milling of Ti–6Al–4V ELI Alloy with the Use of Statistical and Soft Computing Techniques”, Measurement, 90: 25–35, (2016).
45. [45] Paçacı B., Çubuk M. K. and Erol S., “Elektrikli Araç Şarj İstasyonu Konumunun Belirlenmesinde Bulanık Mantık DEMATEL-CBS-TOPSIS Yaklaşımı,” Politeknik Dergisi, 29(2):290209:1-17, (2026).
46. [46] Tezcan B., Eren T., “Forest fire risk assessment in Balikesir using pythagorean fuzzy AHP and pythagorean fuzzy TOPSIS”, Politeknik Dergisi, 29(1):290104:1-13, (2026).