An Integrated Restoration Methodology Based on Adaptive Failure Feature Identification

Yan He,Chuanpeng Hao,Yufeng Li,Yulin Wang,Yan Wang,Wen Ma

DOI: https://doi.org/10.1016/j.rcim.2022.102512

IF: 10.103

2023-01-01

Robotics and Computer-Integrated Manufacturing

Abstract:Remanufacturing is an emerging eco-friendly industry because it consumes less energy, cost, and material to manufacture like-new parts with a warranty to match. However, restoration processes are ad-hoc and complex because the "raw" materials for remanufacturing are returned used parts, which exhibit significant uncertainties in failure features involving failure location, failure mode, failure volume, and failure degree. Thus, customized remanufacturing process planning (RPP) and restoration tool paths should be generated to restore the defects for each part. An integrated restoration methodology based on adaptive failure feature identification for remanu-facturing is proposed to enable efficient and cost-effective remanufacturing. In this study, an adaptive failure feature identification algorithm is developed to identify the failure features on defective parts quickly. In this stage, the point clouds of the nominal model and defective model are used to extract defective regions through Boolean operations and then calculate the failure volume and degree. Based on the identified failure features, a knowledge reuse algorithm is proposed to retrieve the optimal RPP rapidly through mixed case-based reasoning (CBR) and rule-based reasoning (RBR). Finally, a tool path generation algorithm of hybrid Subtractive Manufacturing (SM) and Additive Manufacturing (AM) for the restoration of identified defects. The proposed methodology is verified by remanufacturing a defective blade with multi-defects and is approved to be flexible and effective.