Composite acoustic hole segmentation by semi-supervised learning for robotic multi-spindle drilling of aero-engine nacelle acoustic liners

被引：0

作者：

Dong, Qingyu ^{[1
]}

Mei, Biao ^{[2
]}

Fu, Yun ^{[3
]}

Yang, Yongtai ^{[2
]}

Zhu, Weidong ^{[1
,4
]}

机构：

[1] Zhejiang Univ, Polytech Inst, Hangzhou 310015, Peoples R China

[2] Chinese Acad Sci, Quanzhou Inst Equipment Mfg, Haixi Inst, Quanzhou 362200, Peoples R China

[3] Xizi Spirit Aerosp Ind Zhejiang Ltd, Hangzhou 311222, Peoples R China

[4] Zhejiang Univ, Sch Mech Engn, Hangzhou 310030, Peoples R China

来源：

COMPOSITES PART A-APPLIED SCIENCE AND MANUFACTURING | 2024年 / 184卷

关键词：

Aero-engine nacelle acoustic liners; Semi-supervised acoustic hole segmentation; Weak-to-strong consistency; Multi-reliability;

D O I：

10.1016/j.compositesa.2024.108295

中图分类号：

T [工业技术];

学科分类号：

08 ;

摘要：

The large-scale tiny acoustic holes densely distributed on acoustic liners are essential in aero-engine noise reduction. Accurate segmentation of those holes is fundamental for a robotic multi-spindle drilling system. This paper introduces a novel semi-supervised segmentation method for acoustic holes on composite acoustic liners. This method uses perturbation consistency loss to ensure output consistency while solving the problem of data volume imbalance. Afterward, a multi-reliability enhancement method and a pseudo-label reliability enhancement module are utilized to enhance the model's robustness and accuracy. The segmentation experiments show that our method is superior to UniMatch and FixMatch. When using only 30% of labeled data, our method achieves an IoU of 96.39%, and the porosity differs only 0.038% from the ground truth, which is better than the fully-supervised segmentation method using all data. Our method can meet the accuracy and efficiency requirements of aero-engine nacelle production in China with only limited labeled data.

引用

页数：14