Laser directed energy deposition (LDED) is one of the most important parts of metal additive manufacturing, which can provide fast building speed, allows for large building volumes, and is suitable for part repair. LDED can manufacture components layer by layer through processes of rapid heating, melting, solidification, and cooling with the laser beam as a heat source. However, deposition quality and repeatability of components produced by LDED are poor because of the complex thermal cycle and processing environment, hindering the spread of this technique. Adaptive control technology (ACT) is consistently considered an effective and potential way to solve the problem. Many studies have focused on LDED and established the relations of process parameters, process signatures, and product qualities, which promote the rapid development of ACT, with the development of monitoring devices and data processing technology. We review and discuss the problems existing in the ACT of LDED.
The rapid growth of electric vehicle (EV) market promotes the mass production of lithium-ion batteries. However, the battery production is subjected to high cost and serious environmental issues. Effective and efficient end-of-use lithium-ion battery (LIB) management should be carried out to enhance sustainable development, following the principles of the triple bottom line and circular economy. From the life cycle perspective, battery refurbishing and material recovery are the two major end-of-use options to recover the value of spent batteries. The refurbishing of spent batteries can extend the battery useful life and make full use of battery remaining functional value. Material recycling can recover the battery materials for a new life cycle. However, there still exist many barriers that should be investigated to ensure the success of end-of-use battery recovery. The review covered the pathways to present a full view of end-of-use battery recovery, identified the key bottlenecks in different dimensions, and discussed the strategies for specific scenarios. Industrial practice and pilot projects associated with the two end-of-use options are summarized. In the end, analysis and research suggestions are provided to facilitate the establishment of a sustainable circular battery recovery system.
24 June 2022 in Measurement Science and Technology
In the manufacturing industry, digital radiography (DR) images are often used to detect internal defects in casting parts. With the development of computer technology, increasingly more researchers use computer algorithms instead of manual inspection. However, traditional computer vision methods are generally not efficient and robust. In this study, we propose a DR image defect detection methodology based on deep learning technology. In order to train and evaluate the deep learning model, we create a casting defect DR image dataset, which includes 18 311 DR images labelled for two types of objects—defects and inclusions. In the methodology, an object detection method baseline named YOLOv3_EfficientNet, which replaces the backbone of YOLOv3_darknet53 with EfficientNet, is used. This operation leads to a significant improvement in the mean average precision value on YOLOv3 and greatly reduces the inference time and storage space. Then, a data enhancement method based on DR image features is used, which can increase the diversity of the clarity and the shapes of defects randomly. To further facilitate the deployment of models on embedded devices with an acceptable accuracy loss range, a depth separable convolution operation is adopted. Regarding the bounding box regression, we perform some relevant research in the training and inference stages of the model, and the accuracy of the model was improved in both stages of them according to the experiments. The experiments proved that every type we adopted could benefit the model’s performance.