基于多源信息融合和集成学习的薄壁件 铣削加工变形误差预测

打开文本图片集

中图分类号：TH17

DOI：10.3969/j.issn.1004-132X.2025.06.013 开放科学（资源服务）标识码（OSID）：

Thin-walled Workpiece Milling Deformation Error Prediction Based on Multi-source Information Fusion and Ensemble Learning

YIN Jia1ZHENG Jian² LIU Yao³* JIA Baoguo1DUAN Xiaorui1 1.AVIC Xi'an Aircraft Industry Group Company Ltd.，Xi'an，710089 2.School of Mechano-Electronic Engineering，Xidian University，Xi'an，710071 3.School of Communications and Information Engineering，Xi'an University of Posts and Telecommunications，Xi'an，710121

Abstract： In practical machining processes，the dimensional accuracy of thin-walled workpiece was significantly afected by multiple factors including cutting forces， forced vibrations，chatter phenomena，geometric characteristics of workpiece and material properties，rendering deformation prediction and control particularly challnging.A multi-source information fusion method for deformation error prediction in thin-walled workpiece miling processes was developed. Machining parameters，vibration signals，and other relevant data were integrated to establish a deformation error prediction model through Stacking ensemble learning methodology， with comprehensive experimental validation performed. Comparative analyses reveal that the constructed model demonstrates superior robustness， higher accuracy，and enhanced practicality when compared with conventional data-driven prediction methods.

Key words： thin-walled workpiece；milling process;； deformation error； multi-source information fusion；ensemble learning

0 引言

薄壁零件广泛应用于电子信息、航空航天等领域，是一类非常重要的典型零部件，但其刚度较低，在铣削加工过程中受切削力、强迫振动、颤振等多方面因素影响极易出现较大的变形误差，导致最终产品加工质量下降。（剩余12550字）