International Journal of Manufacturing Research (5 papers in press)

Regular Issues

• A research on metal sealing ring forming process database system and its application  
  by Chang Fan, Wuyang Sun, Zhao Zhang, Dinghua Zhang, Zaifei Zhou, Ming Luo 
  Abstract: Aiming at the forming process of metal sealing ring, in order to enhance and strengthen the automation level of the forming process and the stability of the forming quality, it is necessary to develop the metal sealing ring forming process database system. In order to accumulate data, knowledge and production status during the forming process, it is necessary to collect and store data such as material properties, workpiece dimensions, process equipment, process parameters, forming quality, and forming performance based on the structure-process-performance hierarchical model. In order to facilitate the storage and query of data, we have developed a B/S-based metal sealing ring forming process database system. This database system integrates data management, materials error-proofing, production operation status analysis, and system management, which helps to improve design level, accelerate process iteration, and improve production efficiency.
  Keywords: metal sealing ring; database; structure-process-performance hierarchical model; data model; data management.
  DOI: 10.1504/IJMR.2024.10064273
   
  
• TIG Weld Defect Prediction from Weld Pool Images using Deep Convolutional Neural Network and Transfer Learning  
  by Rachna Verma, Arvind Kumar Verma 
  Abstract: TIG welding is widely used in fabrication, but its quality depends on precise control of welding parameters. This study employs convolutional neural networks (CNNs) and transfer learning to predict welding defects from weld pool images. Six pre-trained CNNs (MobileNet, MobileNetV2, NasNetMobile, InceptionV3, ResNet50V2, and EfficientnetB0) are evaluated for their accuracy and real-time processing ability for a two-class problem (defective vs. non-defective welds) and a six-class problem (classifying good weld, burn through weld defect, contamination weld defect, lack of fusion weld defect, lack of shielding gas weld defect, and high travel speed weld defect). All the models except EfficientnetB0 achieved a very high accuracy. However, based on the inference time and memory size of the models, MobileNetV2 with 99.94% accuracy is recommended for developing an automated TIG welding systems, enabling real-time adjustment of parameters based on weld pool appearance, ensuring high-quality defect-free welds.
  Keywords: TIG welding automation; weld defects; machine learning in TIG; pre-trained networks; CNN in TIG.
  DOI: 10.1504/IJMR.2024.10064357
   
  
• Influence of Cutting Parameters in Hard Turning 40X Steel with Self-Driven Rotary Tool on Surface Roughness using Genetic Programming Method and Artificial Ecosystem-based Optimisation  
  by Trung Nguyen Van, Bien Duong Xuan, Duong Dao Van, Dieu Hoang Thi 
  Abstract: This paper focuses on developing a roughness prediction model based on Genetic Programming (GP) method and evaluates the influence of cutting parameters (CP) on surface roughness (SR) of 40X steel after heat treatment in rotary tool hard turning process. Different GP models are considered and the best model is selected for comparison with the multi-variables regression analysis (MRA) model. Next, the optimal value of CP and their influence on SR are determined through artificial ecosystem-based optimisation algorithm. Two best models GP and MRA were used to investigate the effect of CP on SR value with R2 index higher than 98%. The error value from GP (MSE = 0.014; MAPE = 4.75%) is much smaller than MRA (MSE = 0.045; MAPE = 8.3%). Furthermore, research results show the superiority of GP over MRA in considering the mutual relationship between the input variables for the objective function.
  Keywords: hard turning; surface roughness; multi-variables regression; genetic programming; GP; artificial ecosystem.
  DOI: 10.1504/IJMR.2024.10064816
   
  
• A novel image-based scheme for automated detection and identification of weld defects  
  by Zheng Wang, Lu Li, Weixin Gao 
  Abstract: A guide line setting method is proposed to accurately extract the region of interest (ROI) of the weld. The slice image is segmented from ROI. After noise reduction of the weld slice images using the mean filtering, a contrast check algorithm is proposed to decide whether to enhance the slice images. Then, the suspected defect region (SDR) is obtained by vision-based density clustering segmentation. SDRs can be identified as defects or noise. In this recognition process, a sparse dictionary learning approach is used. After we get the dictionary, we need to solve the corresponding coefficients for each SDR to be tested. The coefficient is obtained by solving the problem of minimising the one-norm of the matrix. From the coefficients, the specific category information of SDR can be directly extracted. Experiments show that when the parameters are properly selected, the sensitivity and specificity of the algorithm are 98.2% and 98.0%.
  Keywords: weld non-destructive testing; image processing; machine vision; coefficient dictionary.
  DOI: 10.1504/IJMR.2024.10065141
   
  
• Autoencoder-Based Defect Detection in PVC Profile Manufacturing  
  by Ahmet Zahit Aslan, Sinan Onal 
  Abstract: This study develops an automatic defect detection system for polyvinyl chloride (PVC) profile manufacturing, addressing inefficiencies in manual inspection. It compares the proposed autoencoder model with other well-known unsupervised deep-learning methods, including GANomaly, f-AnoGAN, and the student-teacher network, for defect detection during extrusion. Utilising a defective PVC profile dataset, the study generates anomaly heat maps through reconstruction errors and assesses model performance using the area under the receiver operating characteristic (ROC) curve. The proposed autoencoder model is found to be optimal for this dataset, offering a balance between efficiency and accuracy. These findings have significant implications for enhancing quality control and reducing defects in PVC manufacturing, with potential applicability in other industrial settings.
  Keywords: automated defect detection; polyvinyl chloride; PVC; profiles; unsupervised deep learning models; autoencoder; quality control in manufacturing.
  DOI: 10.1504/IJMR.2024.10065426