METHOD AND APPARATUS FOR PREDICTING THE OCCURRENCE AND TYPE OF DEFECTS IN AN ADDITIVE MANUFACTURING PROCESS

摘要

A method and an apparatus (100), comprising of a data capture unit (200), a training unit (101), an application unit (102), and an analysis unit (103), for predicting defects (F) in an additive manufacturing process, wherein:- training data and production data, comprising of information on individual layers (L₁,...,L_n) of the additive manufacturing process, are provided,- a classifying function (C) is trained by means of a machine learning technique and the training data (1) to identify pre-warning indicators (W) of defects in printed layers,- the classifying function is applied to production data (2) of the additive manufacturing process,- the classifying function predicts defects (F) in the present or at least one of the subsequent layers.