復(fù)雜空心渦輪葉片點(diǎn)云快速配準(zhǔn)

打開(kāi)文本圖片集

Fast registration of point cloud of complex hollow turbine blade

GOU Chen 1 ，LIAO Xiaobo 1* ，LI Tong1， ZHOU Xin ， ZHUANG Jian 2 ， CAI Yong 1，3

（1. School ofManufacturing Science and Engineering， Southwest University of Science and Technology，Mianyang 621olo，China; 2. School of Mechanical Engineering， Xi'an Jiao Tong University， Xi'an 71OO49， China; 3. Sichuan Electronic and Mechanic Vocational College，Mianyang 62lolO， China） * Corresponding author， E -mail： liaoxiaobo@swust.edu.cn

Abstract： Turbine blades are characterized by their intricate geometries，specialized materials，and com plex manufacturing processes. Three-dimensional models constructed through point cloud alignment serve as critical tools for the comparative analysis and evaluation of blade manufacturing accuracy and quality. However，accurately identifying overlapping regions within point clouds during the alignment process presents significant chalenges，compounded by complex calculations. To address these issues，a novel fast point cloud alignment method for complex hollow turbine blades was developed. This method extracted multi-level features from both source and target point clouds using deep learning techniques and facilitates theexchange of information between these features.Consequently，the global characteristics of the two point clouds could be aligned to focus on corresponding regions without the requirement for an attention mechanism. Experimental results indicate that the root mean square error （RMSE） for both the rotation RMSE （r） and translation RMSE（t） components in the ModelNet4O dataset alignments is reduced by （204號(hào) 34% and 15% ， respectively，compared to the previously established deep learning network PANet. Furthermore，continued training on turbine blade point clouds derived from the ModelNet4O dataset yielded RMSE （r） and RMSE（t） reductions of 83% and 46% ，respectively. This method holds substantial prom ise for enhancing the evaluation processes related to the accuracy of turbine blade manufacturing in future applications.

Key words： point cloud registration; turbine blade; signature information； feature interaction； overlap ping regions

1引言

復(fù)雜空心渦輪葉片是飛機(jī)發(fā)動(dòng)機(jī)和汽輪機(jī)組的核心部件，具有形狀復(fù)雜（壁薄、葉身扭曲大）材料特殊、加工工藝復(fù)雜，以及制造成本高[1]等特點(diǎn)。（剩余11932字）