Zhang S, Hu S, Wang Z (2016) Weld penetration sensing in pulsed gas tungsten arc welding based on arc voltage. Lv N, Xu Y, Li S, Yu X, Chen S (2017) Automated control of welding penetration based on audio sensing technology.
Zhang Z, Wen G, Chen S (2019) Weld image deep learning-based on-line defects detection using convolutional neural networks for Al alloy in robotic arc welding. Jiao W, Wang Q, Cheng Y, Zhang YM (2021) End-to-end prediction of weld penetration: a deep learning and transfer learning based method. Nomura K, Fukushima K, Matsumura T, Asai S (2021) Burn-through prediction and weld depth estimation by deep learning model monitoring the molten pool in gas metal arc welding with gap fluctuation. Liu Y, Zhang Y (2014) Dynamic control of 3D weld pool surface based on human response model. Peng G, Chang B, Wang G, Gao Y, Hou R, Wang S, Du D (2021) Vision sensing and feedback control of weld penetration in helium arc welding process. Li C, Wang Q, Jiao W, M Johnson, Zhang YM (2020) Deep learning-based detection of penetration from weld pool reflection images. Finally, the GTAW penetration monitoring system was designed for online evaluation. The model was visualized by gradient-weighted class activation mapping (Grad-CAM), showing key areas the model depends on to recognize the weld penetration state. The recognition accuracy on the validation set achieved 99.88%, and the recognition time of a single image was about 65 ms. Compared with the MobileNetV2 trained from scratch and the ResNet based on transfer learning, the results show that the proposed method improves model development efficiency on small datasets, while greatly reducing the memory occupied on industrial equipment. Pretrained on the ImageNet dataset, the MobileNetV2-based transfer learning model was used to fit the custom dataset and recognize weld penetration states, including lack of fusion, lack of penetration, desirable penetration, and excessive penetration.
In the study, a flexible visual sensing welding system was developed to construct a gas tungsten arc welding (GTAW) pool image dataset. Weld pool images can be used as the input of convolutional neural networks (CNNs) to recognize the weld penetration state, but it is difficult to design and train an efficient CNN model from scratch for real-time recognition. Weld penetration recognition is of great significance to improve welding quality and automation level.
0 kommentar(er)
