Open AccessMACHINE DATA-BASED PREDICTION OF BLISK BLADE GEOMETRY CHARACTERISTICS
Author(s) -
Alexander Ernst,
Matthias Weigold
Publication year - 2021
Publication title -
mm science journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.195
H-Index - 10
eISSN - 1805-0476
pISSN - 1803-1269
DOI - 10.17973/mmsj.2021_11_2021151
Subject(s) - coordinate measuring machine , machining , computer science , machine tool , mechanical engineering , process (computing) , component (thermodynamics) , aviation , field (mathematics) , engineering drawing , engineering , mathematics , physics , pure mathematics , thermodynamics , aerospace engineering , operating system
The increasing availability of data recording solutions in the field of machining in combination with major developments in Machine Learning and Artificial Intelligence enable new approaches towards optimization in the industrial environment. In the aviation industry, critical components must fulfil extremely high quality standards. This requires a stable and error-free manufacturing process, as well as an extensive geometrical compliance, what is until now verified by long-lasting coordinate measuring machine (CMM) inspection. This publication shows how machine data analysis can contribute to reduce CMM measurement effort and thus decrease component cycle time. For this purpose, production machine data from an aircraft engine Inconel compressor blisk blade 5-axis milling operation was recorded and analysed by subsequent application of machine learning algorithms to predict the geometric measurement characteristics.