Classification of blade’s leading edge based on neural networks in adaptive machining of near-net-shaped blade

Yin, Zikai; Ren, Junxue; Liang, Yongshou

doi:10.1007/s12541-021-00586-y

Cited by 1 publication

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, the current research on the modeling of large curvature inexact forming regions mainly focuses on simple geometric element fitting but does not fully consider the relationship between the design intention of the blade, the similarity relationship between the actual blade surface and theoretical surface, and the complex deformation of blank materials. Our previous manual works addressed the aforementioned issue to a certain extent [3]. For this article, we further proposed an algorithm based on deep learning to accelerate this process and reduce errors caused by human experiences.…”

Section: Related Workmentioning

confidence: 99%

“…This process, however, is time-consuming and depends on the human experience. Deep learning has a cutting-edge advantage in improving efficiency and avoiding human error, based on which we proposed a model reconstruction framework in [3].…”

Section: Introductionmentioning

confidence: 99%

“…Our method can be described in the following steps. In the first stage, we use Generative Adversarial Networks (GAN) [4] to classify optimal reconstructed leading/trailing edge's images based on our previous manual works [3]. However, the reconstructed curves in images cannot be adopted in computeraided design (CAD) software for model reconstruction directly.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

LETR: An End-to-End Detector of Reconstruction Area in Blades Adaptive Machining with Transformer

et al. 2022

Self Cite

View full text Add to dashboard Cite

In the leading/trailing edge’s adaptive machining of the near-net-shaped blade, a small portion of the theoretical part, called the reconstruction area, is retained for securing aerodynamic performance by manual work. The next work is to recognize the reconstruction area of the reconstructed leading/trailing edge’s image. To accelerate this process, an anchor-free neural network model based on Transformer was proposed, named Leading/trailing Edge Transformer (LETR). LETR extracts image features from an aspect of mixed frequency and channel domain. We also integrated LETR with the newest meta-Acon activation function. We tested our model on the self-made dataset LDEG2021 on a single GPU and got an mAP of 91.9%, which surpassed our baseline model, Deformable DETR, by 1.1%. Furthermore, we modified LETR’s convolution layer and named the new model after Ghost Leading/trailing Edge Transformer (GLETR) as a lightweight model for real-time detection. It is proved that GLETR has fewer weight parameters and converges faster than LETR with an acceptable decrease in mAP (0.1%) by test results. The proposed models provide the basis for subsequent parameter extraction work in the reconstruction area.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%