Defects Detection at Additive Manufacturing by Convolutional Deep Learning

Reza Lotfinejad; Asghar Zajkani

Makale detayları

Defects Detection at Additive Manufacturing by Convolutional Deep Learning

Yazarlar:

Reza Lotfinejad | Asghar Zajkani

Yayıncı:

Duzce University

Ana Başlıklar:

Mechanical Engineering (Other)

Dergi:

Düzce University Journal of Technical Sciences Vol.2 Issue 1, pp.36-48

Belge Tipi:

Makale

Yayın Tarihi:

6/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001319861

pISSN:

3023-591X

Anahtar kelimeler:

Özet:

 Additive Manufacturing technologies present a wide array of benefits, including the capacity to manufacture components with complex geometric forms, reduced production expenses, minimized material usage, and time efficiency. This research constitutes a significant effort to pinpoint geometric defects and dimensional irregularities as well as surface quality imperfections in the Fused Deposition Modeling process through the development of a deep learning model utilizing multi-scale convolutional neural networks. The proposed methodology encompasses three distinct scales, each capable of identifying defects of varying dimensions. The model underwent extensive hybridizing procedures for precisely training through diverse datasets, and the training process is repeated numerous times until the desired level of accuracy was attained. A sufficiently extensive image datasets are employed to train the models, leading to the precise calibration of the network. As a result, the necessity for prolonged time and intricate computations to identify large-scale defects is eliminated. The highest validation accuracy for defect detection in this study reached 94.

 Additive Manufacturing technologies present a wide array of benefits, including the capacity to manufacture components with complex geometric forms, reduced production expenses, minimized material usage, and time efficiency. This research constitutes a significant effort to pinpoint geometric defects and dimensional irregularities as well as surface quality imperfections in the Fused Deposition Modeling process through the development of a deep learning model utilizing multi-scale convolutional neural networks. The proposed methodology encompasses three distinct scales, each capable of identifying defects of varying dimensions. The model underwent extensive hybridizing procedures for precisely training through diverse datasets, and the training process is repeated numerous times until the desired level of accuracy was attained. A sufficiently extensive image datasets are employed to train the models, leading to the precise calibration of the network. As a result, the necessity for prolonged time and intricate computations to identify large-scale defects is eliminated. The highest validation accuracy for defect detection in this study reached 94.

 Additive Manufacturing technologies present a wide array of benefits, including the capacity to manufacture components with complex geometric forms, reduced production expenses, minimized material usage, and time efficiency. This research constitutes a significant effort to pinpoint geometric defects and dimensional irregularities as well as surface quality imperfections in the Fused Deposition Modeling process through the development of a deep learning model utilizing multi-scale convolutional neural networks. The proposed methodology encompasses three distinct scales, each capable of identifying defects of varying dimensions. The model underwent extensive hybridizing procedures for precisely training through diverse datasets, and the training process is repeated numerous times until the desired level of accuracy was attained. A sufficiently extensive image datasets are employed to train the models, leading to the precise calibration of the network. As a result, the necessity for prolonged time and intricate computations to identify large-scale defects is eliminated. The highest validation accuracy for defect detection in this study reached 94.

 Eklemeli Üretim teknolojileri, karmaşık geometrik formlara sahip bileşenleri üretme kapasitesi, azaltılmış üretim giderleri, minimum malzeme kullanımı ve zaman verimliliği dahil olmak üzere çok çeşitli avantajlar sunar. Bu araştırma, çok ölçekli evrişimsel sinir ağlarını kullanan bir derin öğrenme modelinin geliştirilmesi yoluyla Erimiş Biriktirme Modelleme sürecindeki geometrik kusurların ve boyutsal düzensizliklerin yanı sıra yüzey kalitesi kusurlarının tespit edilmesi için önemli bir çaba oluşturmaktadır. Önerilen metodoloji, her biri farklı boyutlardaki kusurları tanımlayabilen üç farklı ölçeği kapsamaktadır. Model, çeşitli veri kümeleri aracılığıyla hassas bir şekilde eğitilebilmesi için kapsamlı hibridizasyon prosedürlerinden geçirildi ve eğitim süreci, istenen doğruluk düzeyine ulaşılıncaya kadar birçok kez tekrarlandı. Modelleri eğitmek için yeterince kapsamlı bir görüntü veri kümeleri kullanılır ve bu da ağın hassas kalibrasyonuna yol açar. Sonuç olarak, büyük ölçekli kusurları tespit etmek için uzun süreye ve karmaşık hesaplamalara duyulan ihtiyaç ortadan kalkar. Bu çalışmada kusur tespiti için en yüksek doğrulama doğruluğu 94e ulaştı.

Makale detayları

Defects Detection at Additive Manufacturing by Convolutional Deep Learning

Yazarlar:

Reza Lotfinejad | Asghar Zajkani

Yayıncı:

Duzce University

Ana Başlıklar:

Mechanical Engineering (Other)

Dergi:

Düzce University Journal of Technical Sciences Vol.2 Issue 1, pp.36-48

Belge Tipi:

Makale

Yayın Tarihi:

6/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001319861

pISSN:

3023-591X

Additive Manufacturing

Fused Deposition Modeling

Multiscale Convolutional Neural Network

Deep Learning

Defect DetectionEklemeli Üretim

Birleştirilmiş Biriktirme Modelleme

Çok Ölçekli Evrişimsel Sinir Ağı

Derin Öğrenme

Kusur Tespiti.

Sayı

Makale detayları

Yazarlar: Reza Lotfinejad | Asghar Zajkani

Yayıncı: Duzce University

Ana Başlıklar: Mechanical Engineering (Other)

Dergi: Düzce University Journal of Technical Sciences Vol.2 Issue 1, pp.36-48

Belge Tipi: Makale

Yayın Tarihi: 6/2024

Belge Dili: İngilizce

Document ID (Org No.): 2001319861

pISSN: 3023-591X

Sayı

Yazarlar:

Reza Lotfinejad | Asghar Zajkani

Yayıncı:

Duzce University

Ana Başlıklar:

Mechanical Engineering (Other)

Dergi:

Düzce University Journal of Technical Sciences Vol.2 Issue 1, pp.36-48

Belge Tipi:

Makale

Yayın Tarihi:

6/2024

Belge Dili:

İngilizce

Document ID (Org No.):

2001319861

pISSN:

3023-591X