Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network
Metal additive manufacturing (AM) is gaining increasing attention from academia and industry due to its unique advantages compared to the traditional manufacturing process. Parts quality inspection is playing a crucial role in the AM industry, which can be adopted for product improvement. However, t...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/2/545 |
| _version_ | 1818623329647460352 |
|---|---|
| author | Wenyuan Cui Yunlu Zhang Xinchang Zhang Lan Li Frank Liou |
| author_facet | Wenyuan Cui Yunlu Zhang Xinchang Zhang Lan Li Frank Liou |
| author_sort | Wenyuan Cui |
| collection | DOAJ |
| description | Metal additive manufacturing (AM) is gaining increasing attention from academia and industry due to its unique advantages compared to the traditional manufacturing process. Parts quality inspection is playing a crucial role in the AM industry, which can be adopted for product improvement. However, the traditional inspection process has relied on manual recognition, which could suffer from low efficiency and potential bias. This study presented a convolutional neural network (CNN) approach toward robust AM quality inspection, such as good quality, crack, gas porosity, and lack of fusion. To obtain the appropriate model, experiments were performed on a series of architectures. Moreover, data augmentation was adopted to deal with data scarcity. L2 regularization (weight decay) and dropout were applied to avoid overfitting. The impact of each strategy was evaluated. The final CNN model achieved an accuracy of 92.1%, and it took 8.01 milliseconds to recognize one image. The CNN model presented here can help in automatic defect recognition in the AM industry. |
| first_indexed | 2024-12-16T18:39:20Z |
| format | Article |
| id | doaj.art-cf58af7e759b402fbde178568d642f80 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-12-16T18:39:20Z |
| publishDate | 2020-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-cf58af7e759b402fbde178568d642f802022-12-21T22:21:06ZengMDPI AGApplied Sciences2076-34172020-01-0110254510.3390/app10020545app10020545Metal Additive Manufacturing Parts Inspection Using Convolutional Neural NetworkWenyuan Cui0Yunlu Zhang1Xinchang Zhang2Lan Li3Frank Liou4Department of Mechanical and Aerospace Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USADepartment of Mechanical and Aerospace Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USADepartment of Mechanical and Aerospace Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USADepartment of Mechanical and Aerospace Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USADepartment of Mechanical and Aerospace Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USAMetal additive manufacturing (AM) is gaining increasing attention from academia and industry due to its unique advantages compared to the traditional manufacturing process. Parts quality inspection is playing a crucial role in the AM industry, which can be adopted for product improvement. However, the traditional inspection process has relied on manual recognition, which could suffer from low efficiency and potential bias. This study presented a convolutional neural network (CNN) approach toward robust AM quality inspection, such as good quality, crack, gas porosity, and lack of fusion. To obtain the appropriate model, experiments were performed on a series of architectures. Moreover, data augmentation was adopted to deal with data scarcity. L2 regularization (weight decay) and dropout were applied to avoid overfitting. The impact of each strategy was evaluated. The final CNN model achieved an accuracy of 92.1%, and it took 8.01 milliseconds to recognize one image. The CNN model presented here can help in automatic defect recognition in the AM industry.https://www.mdpi.com/2076-3417/10/2/545additive manufacturing (am)metal defectsquality inspectiondeep learningconvolutional neural network (cnn)defect classification |
| spellingShingle | Wenyuan Cui Yunlu Zhang Xinchang Zhang Lan Li Frank Liou Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network Applied Sciences additive manufacturing (am) metal defects quality inspection deep learning convolutional neural network (cnn) defect classification |
| title | Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network |
| title_full | Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network |
| title_fullStr | Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network |
| title_full_unstemmed | Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network |
| title_short | Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network |
| title_sort | metal additive manufacturing parts inspection using convolutional neural network |
| topic | additive manufacturing (am) metal defects quality inspection deep learning convolutional neural network (cnn) defect classification |
| url | https://www.mdpi.com/2076-3417/10/2/545 |
| work_keys_str_mv | AT wenyuancui metaladditivemanufacturingpartsinspectionusingconvolutionalneuralnetwork AT yunluzhang metaladditivemanufacturingpartsinspectionusingconvolutionalneuralnetwork AT xinchangzhang metaladditivemanufacturingpartsinspectionusingconvolutionalneuralnetwork AT lanli metaladditivemanufacturingpartsinspectionusingconvolutionalneuralnetwork AT frankliou metaladditivemanufacturingpartsinspectionusingconvolutionalneuralnetwork |